Original paper
Arachidonic acid metabolism in TNF, CD178 (CD95 ligand) and ceramide-mediated cytotoxicity

Introduction
TNF (tumour necrosis factor) is a potent mediator of a variety of immune reactions leading to elimination of damaged, infected and tumour cells. TNF-mediated cytotoxicity mostly depends on the induction of pro-apoptotic caspases [1]. However, activation of cytosolic phospholipase A2 (cPLA2) releasing arachidonic acid (AA) from membrane phospholipids has also been reported to play a role in TNF-induced cytotoxicity [2]. AA may subsequently be converted into active lipid mediators, such as leukotrienes and prostanoids by lipoxygenases or cyclooxygenases, respectively [3]. Interestingly, the pro-apoptotic activity of another AA-liberating enzyme, calcium-independent phospholipase A2 (iPLA2), remains poorly understood [4, 5]. The question whether the mechanisms of TNF-mediated cytotoxicity are related to a direct effect of AA and/or its metabolites still awaits elucidation. Some reports have suggested a role of lipoxygenase (LOX) metabolites in TNF cytotoxicity [6, 7, 8], while others have demonstrated direct pro-apoptotic activity of AA [9, 10, 11]. Mechanisms leading to cPLA2 activation in cells transducing cytotoxic signals also remain unclear. Activation of cPLA2 in response to TNF has been proposed to be a consequence of CAPK (ceramide- -activated protein kinase) and MAPK (mitogen- -activated protein kinase) activity [12]. On the other hand, some recent reports suggest that TNF-induced cPLA2 activation does not require ceramide generation. It seems to involve a TNF receptor which may activate the MAP kinase pathway directly through its adapter proteins [12]. CD178 (CD95 ligand, FAS ligand, APO-1 ligand) shares some functional similarities with TNF, such as an ability to induce signal transduction pathways leading to cell death [13]. However, the role of cPLA2 and arachidonic acid in CD178-mediated cytotoxicity remains unclear. In CD95 expressing L929 cells, the CD95-mediated cytotoxicity was reported to be independent of cPLA2 activity [14, 15]. The level of cPLA2 in these cells did not correlate with sensitivity to anti-human CD95 antibodies [15]. Similarly, activation of cPLA2 was insufficient for CD95- -dependent apoptosis of HuT78 lymphoma cells [16]. However, other reports concluded that cPLA2 plays a role in CD95-induced apoptosis of L929 cells [17] as well as in MCF7 human breast carcinoma cells [18], and mouse Sertoli cell line [19]. Considering the above, the aim of the present study was to investigate the role of arachidonic acid metabolism in CD95-mediated cytotoxicity as compared to TNF. Furthermore, a possible role of ceramide in the induction of arachidonic acid- -dependent cytotoxicity was also examined.
Material and methods Target cells lines
A9 cell line (ATCC CCL-1.4), a derivative of L929, stably transfected with pEF-BOS expression vector encoding human CD95 (A9-CD95) as well as the mock-transfected A9 cells were kindly provided by Dr S. Smola-Hess (Cologne, Germany). Both A9 cell lines were cultured in RPMI-1640 (Invitrogen, Carlsbad, CA) containing 10% FCS, penicillin (100 units/ml), streptomycin (100 µg/ml) and amphotericin B (250 ng/ml). Prior to experiments A9 transfectants were expanded for at least two weeks in media containing 800 µg/ml hygromycin B (Calbiochem, San Diego, USA). At least two weeks prior to experiments the selecting antibiotic was withdrawn from the culture media.
Reagents
Recombinant TNF produced in Escherichia coli was from BD Pharmingen (Belgium). Anti-CD95 antagonistic antibody (IgG clone ZB4) and anti-CD95 agonistic antibody (IgM clone CH-11) were obtained from Immnotech (France). C6-ceramide, nordihydroguaiaretic acid (NDGA), indomethacin (INDO) and ActD (actinomycin D) were purchased from Sigma. AACOCF3 (arachidonyl trifluoromethyl ketone) and PACOCF3 (palmitoyl trifluoromethyl ketone) were from Calbiochem (San Diego, USA). Serving as a source of soluble CD178 was cell-free supernatant collected from BHK-21 cells (ATCC CCL-10) stably expressing human CD178 (BHK/CD178 conditioned medium). Serving as controls were supernatants from BHK-21 mock transfectants. These cell lines were kindly provided by Dr S. Smola-Hess (Cologne, Germany). Cytotoxicity assay and assessment of specificity of cytotoxicity mediators For cytotoxicity assays, the cells were seeded in 96-well plates (104 cells/well) and allowed to attach overnight. Next, the cells were preincubated for 30 minutes at 37°C with AACOCF3, PACOCF3 and indomethacin at concentrations ranging from 5 to 50 µM and with NDGA at concentrations ranging from 1 to 10 µM. These amounts were found to be the highest concentrations that were not cytotoxic for the cells. Then the cells were exposed either to TNF or CH-11 antibody or BHK/CD178 conditioned medium or C6-ceramide for 16 h in the absence or presence of 1 µg/ml ActD. Following incubation, viable cells were stained for 15 min at room temperature with 1% crystal violet solution in 30% methanol. Dye uptake was measured by the absorbance at 570 nm in a microtiterplate reader [20]. In experiments with inhibitors TNF was added to a final concentration of 1 ng/ml and CH-11 antibody was used at final concentration of 0.5 µg/ml. The above doses were chosen as they were found to induce 50% cytotoxicity of respective cells. BHK/CD178 conditioned medium was used at a concentration of 1 U/ml that corresponded to a dilution resulting in half-maximal killing of A9-CD95 cells. Exposure of A9 cells to increasing concentrations of C6-ceramide (10-40 µM) led to dose-dependent cell death (data not shown). For subsequent experiments the 40 µM C6-ceramide concentration was used as this concentration led to almost 80% cytotoxicity. Addition of TNF to A9 or A9-CD95 cells resulted in a specific dose-dependent cytotoxicity (data not shown). Both cell lines required co-exposure of TNF and ActD, an inhibitor of RNA synthesis, for induction of cytotoxicity. Only A9-CD95 cells were killed in a dose-dependent manner either by CH-11 antibody or BHK/CD178 conditioned medium. This effect was completely blocked by ZB4 antibody, an anti-CD95 antagonistic antibody (data not shown). Supernatant collected from BHK-21 mock transfectants did not influence the viability of A9-CD95 cells (data not shown).
Statistical analysis
Statistical differences between groups were evaluated by the nonparametric Wilcoxon’s test. All results are expressed as mean ± standard deviation (SD).
Results Effects of inhibitors of arachidonic acid cascade on CD178-mediated cytotoxicity
As shown in Figure 1, cytotoxic activity of TNF was blocked by addition of AACOCF3 and PACOCF3 or by NDGA at their maximal non-toxic concentrations. Effects of AACOCF3 and NDGA were dose-dependent (data not shown). An inhibitory effect of PACOCF3 was found at a concentration of 50 µM (p<0.05 by Wilcoxon’s matched-pair test). Lower doses of PACOCF3 did not affect TNF cytotoxicity. Maximal concentration of indomethacin, a cyclooxygenase inhibitor, did not influence the TNF-mediated cytotoxicity (Figure 1). Contrary to TNF, the cytotoxic effect induced by either CH-11 antibody or BHK/CD178 conditioned medium on A9-CD95 cells was not influenced by any of the tested inhibitors of arachidonic acid metabolism (Figure 1), irrespectively of the concentration applied. Lack of sensitivity of A9-CD95 cells to CD95- -mediated cytotoxicity was not affected by addition of ActD.
Effects of inhibitors of arachidonic acid cascade on C6-ceramide mediated cytotoxicity
AACOCF3, PACOCF3 and indomethacin could not prevent the C6-ceramide-mediated cell death of A9 cells, even when the maximal 50 µM concentration of inhibitors was used. However, pretreatment of the cells with 10 µM NDGA led to slight but reproducible inhibition of C6-ceramide induced cytotoxicity. As shown in Figure 1, NDGA could reduce the C6-ceramide mediated cytotoxicity up to 20%. This effect, however, was not statistically significant.
Discussion
The results of our study support previous findings that AACOCF3 inhibits TNF-mediated cytotoxicity [21]. Inhibition of TNF-mediated cytotoxicity was also observed at the highest non-toxic dose of PACOCF3. To our best knowledge the effect of PACOCF3 on susceptibility to TNF cytotoxicity has not been studied so far. AACOCF3 is a potent inhibitor of cPLA2, which at higher concentration was also shown to inhibit the calcium-independent PLA2 (iPLA2). In contrast, PACOCF3 is considered to be predominantly an inhibitor of iPLA2. At higher concentration, however, it also may inhibit cPLA2 [22]. PACOCF3- -mediated inhibition of TNF was visible only at the highest non-toxic dose, having no effect at lower concentrations. However, discrimination between effects mediated by cPLA2 and iPLA2 would require further studies with highly specific inhibitors. Contrary to TNF, CD178-mediated cytotoxicity was inhibited by neither AACOCF3 nor PACOCF3, thus suggesting that CD178-mediated cell death is both cPLA2- and iPLA2-independent. This observation supports the study of Enari et al. [15], which demonstrated that the cross-linking CD95 receptor induces cell death equally well in both c PLA2- -deficient and -expressing L929 cells. On the other hand, our results are inconsistent with previous results showing that CD95-mediated cytotoxicity of U937 cells depends on iPLA2 activation and AA release [23]. This difference might be explained by promonocytic origin of U937 cells and may reflect the ability of monocytes/macrophages to participate in induction of a proinflammatory response. TNF-mediated cytotoxicity against A9 cells was also inhibited by NDGA, which is consistent with several previous observations [8, 24, 25]. NDGA was initially used as a nonselective LOX inhibitor with antioxidative properties. However, subsequent studies did not support a role for LOX or their products in TNF-induced cell death [24, 26]. This suggests that the effect of NDGA on TNF-induced cytotoxicity may be attributable instead to its anti-oxidant activity. In addition, NDGA effects may also be dependent on inhibition of cPLA2 activation due to its ability to influence calcium signalling [27]. NDGA was also previously shown to inhibit CD95- -mediated apoptosis in human glioma cells [28]. However, NDGA did not affect CD95-mediated cytotoxicity in L929 cells [14]. Our present study supports the latter observation, thus implying that neither AA nor its metabolites are involved in cytotoxicity mediated by CD178. Similarly, one could speculate that neither of the reactive oxygen species seems to be involved in CD95-induced cell death in A9 cells. Both TNF and CD178 are known to activate sphingomyelinases leading to ceramide generation, which are involved in cell cytotoxicity. Indeed, C6-ceramide, a synthetic analogue of natural ceramides, exerted cytotoxic effects on A9 cells. This effect, however, was not affected by any of the tested inhibitors, thus suggesting that signalling pathways activated by ceramides are either independent or are located downstream from PLA2 or COX (cyclooxygenase). This is in agreement with the currently proposed model for c PLA2 activation, which involves its posphorylation by ERK1/2 activated directly by the TNF receptor regardless of ceramide generation [29].
Conclusions
TNF-induced cytotoxicity is cPLA2-dependent. At the same time, CD95- as well as C6 ceramide- -induced cell death does not depend on arachidonic acid metabolism.
Acknowledgements
MO has been awarded with a Young Scientist Stipend from the Foundation for Polish Science 2005, 2006. This paper was supported by Warsaw Medical University Grant No. 1M15/WB6/2005, 1M15/WB3/2005, 1M15/WB2/2006.
References
1. MacEwan DJ. TNF ligands and receptors – a matter of life and death. Br J Pharmacol 2002; 135: 855-75. 2. Cummings BS, McHowat J, Schnellmann RG. Phospholipase A2 in cell injury and death. J Pharmacol Exp Ther 2000; 294: 793-9. 3. Leslie CC. Properties and regulation of cytosolic phospholipase A2. J Biol Chem 1997; 272: 16709-12. 4. Balsinde J, Dennis EA. Bromoenol lactone inhibits magnesium-dependent phosphatidate phosphohydrolase and blocks triacylglycerol biosynthesis in mouse P388D1 macrophages. J Biol Chem 1996; 271: 31937-41. 5. Ma Z, Ramanadham S, Kempe K, Chi XS, Ladenson J, Turk J. Pancreatic islets express a calcium-independent phospholipase A2 enzyme that contains a repeated structural motif homologous to the integral membrane binding domain of ankyrin. J Biol Chem 1997; 272: 11118-27. 6. Haliday EM, Ramesha CS, Ringold G. TNF induces c-fos via a novel pathway requiring conversion of arachidonic acid to a lipoxygenase metabolite. EMBO J 1991; 10: 109-15. 7. O’Donnel VB, Spycher S, Azzi A. Involvement of oxidants and oxidant-generating enzyme(s) in tumour-necrosis-factor-alpha-mediated apoptosis: role for lipoxygenase pathway but not mitochondrial respiratory chain. Biochem J 1995; 310: 133-41. 8. Powell CB, Scott JH, Collins J. Characterization of the protein synthesis independent TNF alpha lytic mechanism in human ovarian and cervical carcinoma cell lines. Gynecol Oncol 1996; 62: 42-8. 9. Wolf LA, Laster SM. Characterization of arachidonic acid-induced apoptosis. Cell Biochem Biophys 1999; 30: 353-68. 10. Scorrano L, Penzo D, Petronilli V, Pagano F, Bernardi P. Arachidonic acid causes cell death through the mitochondrial permeability transition. Implications for tumor necrosis factor-alpha signaling. J Biol Chem 2001; 276: 12035-40. 11. Penzo D, Petronilli V, Angelin A, et al. Arachidonic acid released by phospholipase A2 activation triggers calcium dependent apoptosis through the mitochondrial pathway. J Biol Chem 2004; 279: 25219-25. 12. Kolesnick RN, Kronke M. Regulation of ceramide production and apoptosis. Annu Rev Physiol 1998; 60: 643-65. 13. Hofmann K. The modular nature of apoptotic signaling molecules. Cell Mol Life Sci 1999; 55: 1113-28. 14. Schulze-Osthoff K, Krammer PH, Droge W. Divergent signaling via APO-1/Fas and the TNF receptor, two homologous molecules involved in physiological cell death. EMBO J 1994; 13: 4587-96. 15. Enari M, Hug H, Hayakawa M, Ito F, Nishimura Y, Nagata S. Different apoptotic pathways mediated by Fas and the tumor-necrosis-factor receptor. Cytosolic phospholipase A2 is not involved in Fas-mediated apoptosis. Eur J Biochem 1996; 236: 533-8. 16. Cifone MG, Roncaioli P, De Maria R, et al. Multiple pathways originate at the Fas/APO-1 (CD95) receptor: sequential involvement of phosphatidylcholine-specific phospholipase C and acidic sphingomyelinase in the propagation of the apoptotic signal. EMBO J 1995; 14: 5859-68. 17. Ballou LR, Laulederkind SJF, Rosloniec EF, Raghow R. Ceramide signaling and the immune response. Biochim Biophys Acta 1996; 1301: 273-87. 18. Jaattela M, Benedict M, Tewari M, Shayman JA, Dixit VM. Bcl-x and Bcl-2 inhibit TNF and Fas-induced apoptosis and activation of phospholipase A2 in breast carcinoma cells. Oncogene 1995; 10: 2297-305. 19. Ulisse S, Cinque, B, Silvano G, et al. Erk-dependent cytosolic phospholipase A2 activity is induced by CD95 ligand cross-linking in the mouse derived Sertoli cell line TM4 and is required to trigger apoptosis in CD95 bearing cells. Cell Death Differ 2000; 7: 916-24. 20. Malejczyk J, Malejczyk M, Kock A, et al. Autocrine growth limitation of human papillomavirus type 16-harboring keratinocytes by constitutively released tumor necrosis factor-alpha. J Immunol 1992; 149: 2702-8. 21. Wissing D, Mouritzen H, Egeblad M, Poirier GG, Jaattela M. Involvement of caspase-dependent activation of cytosolic phospholipase A2 in tumor necrosis factor-induced apoptosis. Proc Natl Acad Sci USA 1997; 94: 5073-7. 22. Ackermann EJ, Conde-Frieboes K, Dennis EA. Inhibition of macrophage Ca(2+)-independent phospholipase A2 by bromoenol lactone and trifluoromethyl ketones. J Biol Chem 1995; 270: 445-50. 23. Atsumi G, Tajima M, Hadano A, Nakatani Y, Murakami M, Kudo I. Fas-induced arachidonic acid release is mediated by calcium-independent phospholipase A2 but not cytosolic phospholipase A2 which undergoes proteolitic inactivation. J Biol Chem 1998; 273: 13870-7. 24. Hepburn A, Boeynaems JM, Fiers W, Dumont JE. Modulation of tumor necrosis factor-alpha cytotoxicity in L929 cells by bacterial toxins, hydrocortisone and inhibitors of arachidonic acid metabolism. Biochem Biophys Res Commun 1987; 149: 815-22. 25. Chang DJ, Ringold GM, Heller RA. Cell killing and induction of manganous superoxide dismutase by tumor necrosis factor-alpha is mediated by lipoxygenase metabolites of arachidonic acid. Biochem Biophys Res Commun 1992; 188: 538-46. 26. Hollenbach PW, Zilli DL, Laster SM. Inhibitors of transcription and translation act synergistically with tumor necrosis factor to cause the activation of phospholipase A2. J Biol Chem 1992; 267: 39-42. 27. Culver AC, Michalowski SM, Maia RC, Laster SM. The anti-apoptotic effects of nordihydroguaiaretic acid: Inhibition of cPLA2 activation during TNF-induced apoptosis arises from inhibition of calcium signaling. Life Sci 2005; 77: 2457-70. 28. Wagenknecht B, Gulbins E, Lang F, Dichgans J, Weller M. Lipoxygenase inhibitors block CD95 ligand-mediated apoptosis of human malignant glioma cells. FEBS Lett 1997; 409: 17-23. 29. Kronke M, Adam-Klages S. Role of caspases in TNF-mediated regulation of cPLA2. FEBS Lett 2003; 531: 18-22.