Protein Tyrosine Kinase p56 lck Is Required for Ceramide-induced but Not Tumor Necrosis Factor-induced Activation of NF- k B, AP-1, JNK, and Apoptosis*

Ceramide has been implicated as an intermediate in the signal transduction of several cytokines including tumor necrosis factor (TNF). Both ceramide and TNF activate a wide variety of cellular responses, including NF- k B, AP-1, JNK, and apoptosis. Whether ceramide transduces these signals through the same mechanism as TNF is not known. In the present study we investi-gated the role of the T cell-specific tyrosine kinase p56 lck in ceramide- and TNF-mediated cellular responses by comparing the responses of Jurkat T cells with JCaM1 cells, isogeneic Lck-deficient T cells. Treatment with ceramide activated NF- k B, degraded I k B a , and induced NF- k B-dependent reporter gene expression in a time-de-pendent manner in Jurkat cells but not in JCaM1 cells, suggesting the critical role of p56 lck kinase. These effects were specific to ceramide, as activation of NF- k B by phorbol 12-myristate 13-acetate, lipopolysaccharide, H 2 O 2 , and TNF was minimally affected. p56 lck was also found to be required for ceramide-induced but not TNF-induced AP-1 activation. Similarly, ceramide activated the protein kinases JNK and mitogen-activated protein kinase kinase in Jurkat cells but not in JCaM1 cells. Ceramide also induced cytotoxicity and activated caspases and reactive oxygen intermediates in Jurkat cells but not in JCaM1 cells. Assay— The c-Jun kinase assay performed by a modified method as described earlier after treatment of TNF or ceramide extracts were prepared by lysing cells in buffer containing 20 m M HEPES, pH 7.4, 2 m M EDTA, 250 m M NaCl, 1% Nonidet P-40, 2 m g/ml leupeptin, 2 m g/ml aprotinin, 1 m M phenylmethylsulfonyl fluoride, 0.5 m g/ml benzamidine, and m M dithiothreitol. m were im-munoprecipitated with m g of antibody for at Immune complexes were by with A/G-Sepharose were washed 400 m and kinase buffer (2 400 m m M pH 7.4, m M m M Kinase assays were performed min GST-Jun-(1–79) (2 m 20 m M m M , m M dithiothreitol, and m g 32 PARP a 85-kDa lck Kinase Assay— p56 lck immunocomplex kinase assay performed after cells cell extracts Reactions SDS-PAGE The p56 lck autophosphorylation band analyzed PhosphorImager ceramide transcription Jurkat These of ceramide differs from that of other inducers.

To explore this question, we used the JCaM1 cell line, a genetic variant of Jurkat deficient in p56 lck protein because of the deletion of exon 7 in p56 lck mRNA (28). p56 lck is a cytoplasmic tyrosine kinase, has a molecular size of 56 kDa, is a member of the Src family that is expressed highly in T cells, and binds to the cytoplasmic domain of the CD4 receptor (29), all of which make it a candidate for the ceramide and TNF pathways. It is required for T cell signaling in the human Jurkat T cell leukemia line (30). This protein-tyrosine kinase mediates NF-B activation upon interaction of the human immunodeficiency virus type 1 envelope glycoprotein gp120 with the CD4 receptor (31). The precise role p56 lck plays in the ceramide or TNF-induced signaling is not known, however. We compared the cellular responses induced by ceramide with those induced by TNF in this cell line. We also used JCaM1 cells that had been reconstituted by transfection with the p56 lck gene (30). The studies indicated that ceramide can activate p56 lck and is required for ceramide-induced activation of NF-B, AP-1, JNK, MAPK kinase, and apoptosis. Although several cellular responses of ceramide mimicked those of TNF, p56 lck was not found to be essential for TNF-induced cellular responses. man TNF, purified to homogeneity with a specific activity of 5 ϫ 10 7 units/mg, was kindly provided by Genentech, Inc. (South San Francisco, CA). Antibody against IB␣, p50, p65, JNK1, c-Jun, c-Fos, Cyclin D1, c-Rel, Lck, and double-stranded oligonucleotide having the AP-1 consensus sequence were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Poly(ADP-ribose) polymerase (PARP) and Phospho-IB␣ (Ser 32 ) antibody were purchased from New England Biolabs. A 10 mM solution of ceramide was made in Me 2 SO and then further diluted in the cell culture medium.
Cell Lines-The cell lines Jurkat (human T cells), and JCaM1 (p56 lckdeficient) were obtained from the American Tissue and Cell Culture Collection (ATCC, Rockville, MD). JCaM1 cells transfected with the p56 lck gene were kindly supplied by Dr. Arthur Weiss (The University of California, San Francisco, CA). The characterization of these cells has been previously reported (30). All cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum and 1ϫ antibiotics-antimycotics.
NF-B Activation Assay-To assay NF-B activation, we prepared nuclear extracts and performed electrophoretic mobility shift assays (EMSA) as described (32).
Western Blot for IB␣-To assay IB␣, postnuclear (cytoplasmic) extracts were prepared (32) from treated cells and resolved on 10% SDS-polyacrylamide gels. After electrophoresis, the proteins were electrotransferred to nitrocellulose filters, probed with rabbit polyclonal antibodies against either phospho-IB␣ or IB␣, and detected by chemiluminescence (ECL, Amersham Pharmacia Biotech).
NF-B-dependent Reporter Gene Transcription-Ceramide-induced, NF-B-dependent reporter gene transcription was measured as described previously (33). Briefly, cells (0.1 ϫ 10 6 cells/well) were plated in 6-well plates and then transfected with plasmid DNA (0.5 g) for NF-B promoter DNA that had been linked to heat-stable secretory alkaline phosphatase (SEAP) by the calcium phosphate method. After 10 h, cells were treated with different concentrations of TNF or ceramide. Twelve hours later, cell culture-conditioned medium was harvested, and 25 l was analyzed for alkaline phosphatase activity essentially as described by the protocol of CLONTECH Inc. (Palo Alto, CA). The activity of SEAP was assayed on a 96-well fluorescent plate reader (Fluoroscan II, Lab Systems) with excitation set at 360 nm and emission at 460 nm. This reporter system was specific, because TNF-induced NF-B SEAP activity was inhibited by overexpression of IB␣ mutants lacking either Ser 32 or Ser 36 (33).
Cytotoxicity Assay-The ceramide-induced cytotoxicity was measured by the modified tetrazolium salt 3-(4 -5-dimethylthiozol-2-yl) 2-5diphenyl-tetrazolium bromide (MTT) assay (34). Briefly, cells (10,000 cells/well) were incubated in the presence or absence of the indicated test sample in a final volume of 0.1 ml for 72 h at 37°C. Thereafter, 0.025 ml of MTT solution (5 mg/ml in phosphate-buffered saline) was added to each well. After a 2-h incubation at 37°C, 0.1 ml of the extraction buffer (20% SDS, 50% dimethyl formamide) was added. After an overnight incubation at 37°C, the optical densities at 590 nm were measured using a 96-well multiscanner autoreader (Dynatech MR 5000), with the extraction buffer as a blank.
Immunoblot Analysis of PARP Degradation-Ceramide-and TNFinduced apoptosis was examined by proteolytic cleavage of PARP (32). Briefly, Jurkat and JCaM1 cells (2 ϫ 10 6 /ml) were activated with different concentrations of TNF or ceramide for 24 h, and then cell extracts were prepared by incubating the cells for 30 min on ice in 0.05 ml of buffer containing 20 mM HEPES, pH 7.4, 2 mM EDTA, 250 mM NaCl, 0.1% Nonidet P-40, 2 g/ml leupeptin, 2 g/ml aprotinin, 1 mM phenylmethylsulfonyl fluoride, 0.5 g/ml benzamidine, and 1 mM dithiothreitol for 30 min. The lysate was centrifuged, and the supernatant was collected. Cell extract protein (50 g) was resolved on 7.5% SDS-PAGE, electrotransferred onto a nitrocellulose membrane, blotted with anti-PARP antibody, and then detected by chemiluminescence (ECL; Amersham Pharmacia Biotech). Apoptosis was represented by the cleavage of 116-kDa PARP into a 85-kDa peptide product.

RESULTS
To study the role of p56 lck in cellular signaling activated by ceramide and TNF, we used JCaM1 cells, which are known to be p56 lck -deficient. Jurkat cells were used as a p56 lck -positive control. Most ceramide-induced cellular responses are similar to those of TNF. To determine how ceramide signaling differs from that of TNF, the two agents were compared throughout this study.
Ceramide Activates NF-B in Jurkat Cells but Not in JCaM1 Cells-Although both TNF and ceramide activate NF-B, whether activation requires p56 lck is not known. Jurkat and JCaM1 cells were treated with various concentrations of either TNF or ceramide for 30 min, and nuclear extracts were prepared and examined for NF-B activation by EMSA (Fig. 1A). TNF (upper panels) activated NF-B in both Jurkat and JCaM1 cells in a dose-dependent manner, with optimum activation (5-fold) at around 100 pM. Ceramide (lower panels) activated NF-B in a dose-dependent manner in Jurkat cells optimally at 5 M (4-fold), but no activation was found in JCaM1 cells. These results suggest that p56 lck kinase is required for ceramide-induced but not for TNF-induced activation. To determine if this effect was time-dependent, Jurkat and JCaM1 cells were treated with either TNF (100 pM) or ceramide (10 M) for different times and then examined for NF-B activation (Fig. 1B). TNF activated NF-B in both cell types with comparable kinetics (upper panels). Ceramide (lower panels) activated NF-B in a time-dependent manner in Jurkat cells with optimum activation occurring at 15 min, but no significant activation of NF-B was observed in p56 lck -deficient JCaM1 cells, even after 60 min.
NF-B Activation Induced by PMA, LPS, H 2 O 2 , Ceramide, and TNF-NF-B is activated by a wide variety of stimuli through a pathway involving overlapping and nonoverlapping steps (36). Thus we sought to examine if p56 lck is required for NF-B activation induced by PMA, LPS, and H 2 O 2 . Jurkat and JCaM1 cells were stimulated with PMA (25 ng/ml), serum activated LPS (1 g/ml), H 2 O 2 (250 M), TNF (0.1 nM), and ceramide (10 M) for 30 min at 37°C. After these treatments, nuclear extracts were prepared and assayed for NF-B by EMSA (Fig. 1C). PMA, LPS, TNF, and H 2 O 2 activated NF-B in both cell types, but again ceramide activated the transcription factor only in Jurkat cells. These results suggest that the mechanism of activation of NF-B by ceramide differs from that of other inducers.
Components of Ceramide-induced NF-B and Its Specificity-Activated NF-B typically consists of p50 and p65 homodimers or heterodimers (37). To determine the composition of the ceramide-induced NF-B complex, we prepared nuclear extracts from untreated or ceramide-treated (10 M) Jurkat cells (2 ϫ 10 6 /ml), incubated then for 15 min with different antibodies and unlabeled NF-B probe, and then assayed them for NF-B by EMSA (Fig. 1D). Both anti-p50 and anti-p65 antibodies supershifted the NF-B complex, whereas irrelevant anti-cyclin D1, anti-cRel, or preimmune serum had no effect on the complex. The NF-B band disappeared by competition with wild-type oligo but not with mutant oligo.
Ceramide Did Not Induce IB␣ Degradation in JCaM1 Cells-NF-B activation by most inducers requires IB␣ degradation (37). Previously it has been shown that NF-B activation induced by UV, pervanadate (PV), or reoxygenation does not coincide with IB␣ degradation (38). Whether p56 lck is required for ceramide-induced IB␣ degradation was also examined ( Fig. 2A). As expected TNF-induced IB␣ degradation reached maximum at 15 min in both Jurkat and JCaM1 cells. The resynthesis of IB␣ occurred at 30 min in Jurkat cells but at 60 min in JCaM1 cells, suggesting that Lck may decrease the rate of resynthesis of TNF-induced IB␣ ( Fig. 2A, upper panels). Like TNF, ceramide induced IB␣ degradation, reaching the maximum at 15 min in Jurkat cells ( Fig. 2A, lower left  panel). In p56 lck -deficient JCaM1 cells, however, no ceramideinduced IB␣ degradation was observed ( Fig. 2A, lower right  panel).
Ceramide Did Not Induce IB␣ Phosphorylation in JCaM1 Cells-That PV activates NF-B without IB␣ phosphorylation in p56 lck -deficient JCaM1 cells has been reported (38). Here we examined what effect p56 lck had on ceramide-induced IB␣ phosphorylation. To detect the phosphorylated form of IB␣ on the Western blot, we used antibodies specific to the serine 32 phosphorylated form of IB␣. Cytoplasmic extracts treated with ceramide or TNF for different times were probed with antibodies against the phosphorylated IB␣ and examined by chemiluminescence. TNF induced IB␣ phosphorylation as early as 5 min in both cell types (Fig. 2B, upper panels), whereas ceramide induced IB␣ phosphorylation in Jurkat cells but not in JCaM1 cells (Fig. 2B, lower panels). Because p56 lck is a protein-tyrosine kinase and IB␣ phosphorylation detected is on serine, p56 lck must regulate an IB␣ kinase that phosphorylates IB␣ directly.
NF-B-dependent Reporter Gene Expression-NF-B binding to the DNA and IB␣ degradation is not sufficient to suggest that p56 lck is required for NF-B-dependent reporter gene expression (39). Therefore, the effect of p56 lck on ceramide-induced reporter gene expression was examined. As shown in Fig. 3, TNF induced reporter gene expression in both Jurkat and JCaM1 cells in a dose-dependent manner. Ceramide induced expression in a dose-dependent manner in Jurkat cells but not in JCaM1 cells, suggesting that p56 lck was also required for ceramide-induced NF-B-mediated reporter gene expression.
AP-1 Activation-Most agents that activate NF-B also activate AP-1. Whether AP-1 activation by either TNF or ceramide requires p56 lck was investigated. To determine the role of p56 lck in AP-1 activation, Jurkat and JCaM1 cells were treated with various concentrations of either TNF or ceramide for 30 min, and the nuclear extracts were prepared and examined for AP-1 activation by EMSA (Fig. 4A). TNF activated AP-1 in both Jurkat and JCaM1 cells in a dose-dependent manner, with optimum activation at around 100 pM (upper panels). Ceramide activated AP-1 in a dose-dependent manner in Jurkat cells, but no activation was found in JCaM1 cells (lower panel). These results suggest that p56 lck kinase is also not required for TNFinduced AP-1 activation but is required for ceramide-induced activation. To determine if this effect is time-dependent, we treated Jurkat and JCaM1 cells with either TNF (100 pM) or ceramide (10 M) for different times and then examined the cells for AP-1 activation (Fig. 4B). TNF activated NF-B in both cell types with similar kinetics. Ceramide activated NF-B in Jurkat cells in a time-dependent manner, but again no significant activation was observed in p56 lck -deficient JCaM1 cells.
Like NF-B, AP-1 can be activated by a wide variety of stimuli (40). To examine if p56 lck is required for AP-1 activation induced by PMA, LPS, and H 2 O 2 , Jurkat and JCaM1 cells were stimulated with PMA (25 ng/ml), serum activated LPS (1 g/ ml), H 2 O 2 (250 M), TNF (0.1 nM), or ceramide (10 M) for 30 min at 37°C. After these treatments, nuclear extracts were prepared and then assayed for AP-1 by EMSA (Fig. 4C). PMA, FIG. 4. Effect of TNF and ceramide on AP-1 activation. A, Jurkat and JCaM1 cells (2 ϫ 10 6 /ml) were stimulated with different concentrations of TNF and ceramide for 30 min. After these treatments, nuclear extracts were prepared and then assayed for AP-1 as described under "Experimental Procedures." B, cells were incubated at 37°C with 0.1 nM TNF and 10 M ceramide for the indicated times. After these treatments nuclear extracts were prepared and then assayed for AP-1. C, effect of PMA, LPS, H 2 O 2 , TNF, and ceramide on AP-1 activation. Jurkat and JCaM1 cells were stimulated with PMA (25 ng/ml), serum activated LPS (1 g/ml), H 2 O 2 (250 M), TNF (0.1 nM), and ceramide (10 M) for 30 min at 37°C. After these treatments nuclear extracts were prepared and then assayed for AP-1. D, supershift and specificity of AP-1. Nuclear extracts were prepared from untreated or ceramide (10 M) -treated Jurkat cells (2 ϫ 10 6 /ml), incubated for 15 min with different antibodies and unlabeled AP-1 probe, and then assayed for AP-1 as described under "Experimental Procedures." LPS, TNF, and H 2 O 2 activated AP-1 in both cell types, but ceramide activated it only in Jurkat cells. These results suggest that the mechanism of activation of AP-1 by ceramide differs from that of other inducers.
To determine the composition of ceramide-induced AP-1 complex, nuclear extracts were prepared from untreated or ceramide-treated (10 M) Jurkat cells (2 ϫ 10 6 /ml), incubated for 15 min with different antibodies and unlabeled AP-1 oligo, and then assayed for AP-1 by EMSA (Fig. 4D). Both anti-c-Fos and anti-c-Jun antibodies supershifted the AP-1 complex, whereas irrelevant anti-cyclinD1, anti-p50, or preimmune serum had no effect on the complex. The AP-1 band disappeared by competition with wild-type oligo.
JNK Activation-The activation of AP-1 requires the activation of a stress-activated protein kinase, JNK (40). To determine the role of p56 lck in JNK activation, Jurkat and JCaM1 cells were treated with various concentrations of either TNF or ceramide for 15 min, and the cell extracts were prepared and examined for JNK activation by immune complex kinase assays (Fig. 5). TNF activated JNK in both Jurkat and JCaM1 cells in a dose-dependent manner, with optimum activation (6-fold) at around 1000 pM concentration (upper panels). Ceramide activated JNK in a dose-dependent manner in Jurkat cells, but no activation was found in JCaM1 cells (lower panel). These results suggest that p56 lck kinase plays no role in TNFinduced JNK activation but it does for ceramide-induced activation.
MAPK Kinase Activation-The activation of JNK and NF-B is regulated by an upstream kinase MAPK kinase or MEK (40,41). To determine the role of p56 lck in MEK activation, we treated Jurkat and JCaM1 cells with various concentrations of either TNF or ceramide for 30 min, prepared the cell extracts, and examined them for MEK activation by Western blot using an antibody that detects the phosphorylated form of MAPK (Fig. 6). TNF activated MEK in both Jurkat and JCaM1 cells in a dose-dependent manner, with optimum activation at around 100 pM concentration (upper panels). Ceramide activated MEK in a dose-dependent manner in Jurkat cells, but no significant activation was found in JCaM1 cells (lower panel). These results suggest that p56 lck kinase plays no significant role in TNF-induced MEK activation, but it does play an important role in ceramide-induced activation.
Apoptosis Induction-Several reports indicate that ceramide can induce apoptosis in various cell types (see Ref. 1). Whether p56 lck is required for the cytotoxic effects of ceramide is not known. To determine the role of p56 lck in cytotoxicity, Jurkat and JCaM1 cells were treated with various concentrations of either TNF or ceramide for 72 h and then examined for cell viability by MTT dye uptake assay (Fig. 7A). TNF induced cytotoxicity in both Jurkat and JCaM1 cells in a dose-dependent manner, the optimum effect occurring around 1 nM concentration (upper panels). Ceramide induced cytotoxicity in a dosedependent manner in Jurkat cells, but no significant cytotoxicity was found in JCaM1 cells (lower panel). These results suggest that p56 lck kinase plays no significant role in TNF-induced cytotoxicity, but it does play an important role for ceramide-induced cytotoxic effects.
The cytotoxic effects of TNF are mediated through the activation of cascade of caspases (see Ref. 42). Caspase-2, -3, -7, and -9 are known to cleave PARP protein substrate. Whether ceramide activates these caspases and whether the activation of these caspases requires p56 lck are not known. To determine FIG. 5. Effect of TNF or ceramide on JNK activation. Jurkat and JCaM1 cells were stimulated with different concentrations of either TNF or ceramide as indicated for 15 min at 37°C. Then the cells were washed, and pellets were extracted and assayed for JNK activation as described under "Experimental Procedures." To demonstrate equal loading, 50 g of protein from the same extract was analyzed in 9% SDS-PAGE to detect JNK1 by Western blot analysis.
FIG. 6. Effect of TNF or ceramide on MAPK kinase activation. Jurkat and JCaM1 cells were stimulated with different concentrations of either TNF or ceramide as indicated for 30 min at 37°C. Then the cells were washed, pellets were extracted, and 50 g of protein was analyzed in 10% SDS-PAGE. Western blot was developed against anti-MAPK phosphorylated antibody (New England Biolabs). To show equal loading, the same blot was stripped and reprobed with extracellular response kinase 2 (ERK2) antibody. the role of p56 lck in caspase activation, we treated Jurkat and JCaM1 cells with various concentrations of either TNF or ceramide for 24 h and then prepared the cellular extracts and examined them for PARP cleavage by Western blot analysis (Fig. 7B). TNF induced PARP cleavage in both Jurkat and JCaM1 cells in a dose-dependent manner, with optimum effect at around 1 nM concentration (upper panels). TNF-induced PARP cleavage was somewhat enhanced in p56 lck -deficient JCaM1 cells. Ceramide induced PARP cleavage in a dose-dependent manner in Jurkat cells, but no significant caspase activation was found in JCaM1 cells (lower panel). These re-sults suggest that p56 lck kinase plays an important role in ceramide-induced activation of apoptosis but perhaps plays little role in TNF-induced apoptosis.
Ceramide-induced p56 lck Activation-From the studies indicated above it is clear that p56 lck -deficient cells are unable to activate NF-B, AP-1, JNK, MEK, and apoptosis induced by ceramide. This implies that ceramide must mediate its effects through activation of p56 lck kinase. We have previously shown that TNF can activate p56 lck kinase (35). Whether ceramide can activate p56 lck is, however, not known. To determine the activation of p56 lck , we first assayed p56 lck protein in Jurkat cells and its absence in JCaM1 cells by Western blot analysis using p56 lck antibodies. As shown in Fig. 8A, p56 lck protein was present in Jurkat cells but not in JCaM1 cells. Then activation of p56 lck was examined by treating cells with different concentrations of either TNF or ceramide for 15 min and then testing them for autophosphorylation of p56 lck (Fig. 8B). TNF induced the autophosphorylation of p56 lck in Jurkat cells but not in JCaM1 cells in a dose-dependent manner (upper panels). Ceramide also activated p56 lck in Jurkat cells but not in JCaM1 cells (lower panel). These results suggest that p56 lck is activated by both TNF and ceramide but is required only for ceramide-mediated cellular responses.
Ceramide-induced Cellular Responses Can Be Reversed by Transfection of p56 lck Gene in JCaM1 Cells-To further confirm the role of p56 lck in ceramide signaling, we used JCaM1 cells that had been reconstituted by transfection of the p56 lck gene (30). The reconstituted cells expressed p56 lck protein (Fig.  8A) and this protein could be activated by TNF and ceramide in a dose-dependent manner (Fig. 9A). We further examined these cells for ceramide-induced NF-B activation (Fig. 9B), JNK activation (Fig. 9C), and cytotoxicity (Fig. 9D). The presence of p56 lck reversed the ceramide-induced activation of NF-B, JNK, and cytotoxicity in a dose-dependent manner, and it had no significant effect on TNF-induced activation. DISCUSSION Although ceramide activates several cellular responses similar to those activated by TNF, whether they are mediated through the same mechanism is not understood. By using p56 lck -deficient cells, in this report we demonstrate that Lck is required for ceramide-induced activation of NF-B, AP-1, JNK, MEK, and apoptosis, but not for that induced by TNF. We found that in control Jurkat cells both TNF and ceramide induced p56 lck activity, but it was needed only for ceramideinduced cellular responses. The transfection of p56 lck -deficient cells with p56 lck gene reconstituted the ceramide-induced cellular responses.
To our knowledge this is the first report to indicate that p56 lck is required for ceramide-mediated activation of NF-B, AP-1, JNK, MEK, and apoptosis. Three pieces of independent evidence in our studies suggest the role of p56 lck in ceramide signaling. First, ceramide signaling is interrupted in p56 lckdeficient cells. Second, ceramide can activate p56 lck kinase activity in cells where signaling is intact. Third, reconstitution of p56 lck -deficient cells with p56 lck gene reverses ceramide signaling. The activation of NF-B by human immunodeficiency virus type 1-derived envelope glycoprotein gp-120 has been shown to require p56 lck (31). The p56 lck has also been shown to be required for NF-B activation induced by reoxygenation and by PV (38). Based on p56 lck -deficient Jurkat variants, the role of p56 lck in PV-induced tyrosine phosphorylation of IB␣ and NF-B activation was suggested (38). Our laboratory and Imbert et al. (38) showed that PV induced phosphorylation of IB␣ at position 42 (43). How tyrosine phosphorylation activates NF-B is not understood. In the present studies we found that ceramide induces serine phosphorylation of IB␣ in Jurkat cells but not in p56 lck -deficient cells, thus suggesting that p56 lck indirectly affects serine phosphorylation of IB␣. Thus p56 lck must modulate the function of IB␣ kinase, which phosphorylates IB␣ (37).
We showed that p56 lck is required for ceramide-induced activation of JNK and MEK. TNF-induced activation of JNK and MEK was found to be p56 lck -independent. Our results are consistent with a previous report that T-cell antigen receptormediated activation of the MAPK pathway requires p56 lck kinase (44). Like us, Denny et al. (44) employed p56 lck -deficient JCaM1 cells to demonstrate the requirement for p56 lck . The activation of JNK by L-selectin was also found to be mediated through p56 lck kinase (45).
How p56 lck mediates the activation of NF-B, AP-1, JNK, MEK, and apoptosis by ceramide is not clear. The activation of these cellular responses requires the generation of reactive oxygen intermediates (37,40,46,47). For instance, overexpression of the antioxidant enzymes superoxide dismutase and ␥-glutamylcysteine synthetase has been shown to suppress the activation of NF-B induced by ceramide (32,48). Thus it is possible that p56 lck -induced reactive oxygen intermediate generation mediates the activation of NF-B induced by ceramide. H 2 O 2 has been shown to activate p56 lck enzyme (49) but whether p56 lck can mediate reactive oxygen intermediate generation has not been reported. These observations also suggest that p56 lck is located upstream in the ceramide signaling pathway. Although we have previously reported that TNF activated p56 lck (35), this is the first report to indicate that ceramide can also activate this kinase. Recently, it was shown by Hanna et . Then cell extracts were prepared, and 800 g of protein was subjected to immunoprecipitation with anti-p56 lck antibody. Kinase assay was performed as described under "Experimental Procedures." Autophosphorylated p56 lck was detected by radioactivity.
al. (7) that both TNF and ceramide activate p21 ras and phosphotidylinositol 3-kinase through activation of a tyrosine kinase activity. Which tyrosine kinase was not identified. It is possible that p56 lck activation by TNF and ceramide activates both p21 ras and phosphotidylinositol 3-kinase.
In our studies we found that although Lck deficiency interrupted the ceramide signaling, TNF-mediated cellular signaling was unaffected, thus suggesting that TNF does not mediate its effects through ceramide. This conclusion is in agreement with reports that show that TNF-induced NF-B and JNK activation is not dependent on ceramide production (12,14,15,20,26). On the contrary, Gamard et al. (26) showed that ceramide treatment of Jurkat cells, the same cell type as used in our studies, blocked phorbol ester-induced NF-B activation. Our results also suggest that TNF and ceramide induce apoptosis through distinct mechanisms. These results are in agreement with those of Guo et al. (17), who showed that in rat mesangial cells TNF induces cell death by a mechanism distinct from that of ceramide. These workers also showed that ceramide kills cells by necrosis, whereas TNF kills by apoptosis (17). That breast tumor MCF-7 cells, which are resistant to TNF, can undergo ceramide-induced apoptosis, also suggests distinct pathways for TNF and ceramide (20). That p56 lck can play a role in apoptosis induced by fas and other agents has been previously reported (50,51). In agreement with our studies, it was shown recently that p56 lck is required for caspase-8 activation and apoptosis in response to ionizing radiation (50). Another group, however, showed that in nontransformed T lymphocytes the p56 lck deficiency induces cell cycle arrest and hypersusceptibility to apoptosis (51). Overall our results demonstrate for the first time that ceramide can activate p56 lck and the latter plays a major role in ceramide signaling but not in TNF signaling. FIG. 9. Cellular responses are reversed after reconstitution of p56 lckdeficient JCaM1 cells with p56 lck . TNF and ceramide induces p56 lck autophosphorylation (A), NF-B activation (B), JNK activation (C), and cytotoxicity (D) in reconstituted JCaM1/p56 lck cells. A, JCaM1/p56 lck cells were stimulated with different concentrations of either TNF or ceramide for 15 min at 37°C. Then cell extracts were prepared, and 800 g of protein was subjected to immunoprecipitation with anti-p56 lck antibody, and then a kinase assay was done as described under "Experimental Procedures." Autophosphorylated p56 lck was detected as a radioactive band. B, JCaM1/p56 lck cells (2 ϫ 10 6 /ml) were stimulated with different concentrations of TNF and ceramide for 30 min. After these treatments, nuclear extracts were prepared and then assayed for NF-B as described under "Experimental Procedures." C, JCaM1/p56 lck cells (2 ϫ 10 6 /ml) were stimulated with different concentrations of TNF and ceramide for 15 min. After these treatments, cell extracts were prepared and then assayed for JNK as described under "Experimental Procedures." D, JCaM1/p56 lck cells (5 ϫ 10 3 /0.1 ml) were treated with different concentrations of TNF and ceramide for 72 h at 37°C in a CO 2 incubator. Relative cell viability was then determined by the MTT method. The results shown are the mean (Ϯ S.E.) optical density of triplicate assays.