Phorbol 12-myristate 13-acetate inhibits death receptor-mediated apoptosis in Jurkat cells by disrupting recruitment of Fas-associated polypeptide with death domain.

Regulation of death receptor-mediated apoptosis is incompletely understood. Previous studies have demonstrated that phorbol 12-myristate 13-acetate (PMA), a protein kinase C activator, inhibits Fas (CD95)-mediated apoptosis in Jurkat (type II) cells but not SKW6.4 (type I) cells. In this study, we demonstrated that PMA also protects Jurkat cells from apoptosis induced by tumor necrosis factor-alpha and the tumor necrosis factor-alpha-related apoptosis-inducing ligand (TRAIL). Interestingly, PMA failed to protect Jurkat cells from apoptosis induced by other agents, including etoposide, camptothecin, and gamma-irradiation. Analysis of the initial events induced by agonistic anti-Fas antibodies revealed that PMA inhibited Fas binding to Fas-associated polypeptide with death domain (FADD) in Jurkat cells but not in SKW6.4 cells. Although the protein kinase inhibitor bisindoylmaleimide VIII increased apoptosis induced by agonistic anti-Fas antibody, tumor necrosis factor-alpha, and TRAIL, these effects were not observed with the protein kinase C inhibitor H7 and were not associated with increased FADD recruitment to Fas. These results indicate that PMA inhibits death signaling induced by a number of discrete receptors and suggest that the effects are mediated at the level of receptor-mediated adaptor molecule recruitment.

Fas (also called CD95/APO-1) is a 45-kDa type I transmembrane protein that is expressed by a variety of cells (1). It belongs to a subgroup of the TNF-␣ 1 receptor/nerve growth factor receptor superfamily that also contains type I TNF-␣ receptor, death receptor 4 (TRAIL receptor 1), and death receptor 5 (TRAIL receptor 2) (2, 3). Each of these polypeptides contains a conserved cytoplasmic domain of ϳ80 amino acids called a death domain that is essential for cell death signaling (2,3).
Fas participates in regulation of the immune system and in the response to certain anticancer drugs. Fas-mediated signaling is required for negative selection of T cells, termination of immune responses after antigen stimulation, and elimination of virally infected cells and tumor cells by cytotoxic lymphocytes (3)(4)(5)(6). In addition, a number of chemotherapeutic agents induce FasL synthesis (7)(8)(9)(10). Experiments utilizing blocking anti-Fas antibodies (11,12) or mice with mutant Fas receptor (12) have demonstrated an essential role for FasL-Fas interactions in 5-fluorouracil-induced cytotoxicity. The role of Fas in the action of other chemotherapeutic agents is less clear (13,14) but is currently under investigation in a number of laboratories.
The Fas signaling pathway has been extensively characterized. Ligation of Fas by FasL or agonistic anti-Fas antibody results in sequential recruitment of FADD and procaspase-8 to the death domain of Fas to form the DISC (5,15). Juxtaposition of procaspase-8 molecules in the DISC is thought to initiate the autocatalytic processing and release of active caspase-8 into the cytoplasm (16). In so-called type I cells, caspase-8 activation at the DISC is sufficient to lead to direct proteolytic activation of caspase-3 (17,18), which then cleaves a number of different substrates to generate the apoptotic phenotype (19). In contrast, in type II cells, caspase-8 generation is not sufficient to activate caspase-3 (17). Instead, caspase-8 cleaves the cytosolic protein Bid to yield a fragment that facilitates translocation of Bax and Bak to the outer mitochondrial membrane (20 -22), where these proapoptotic Bcl-2 family members induce cytochrome c release (23,24) and subsequent triggering of apoptotic changes through the activation of caspase-9 (17,25).
The events occurring after ligation of other death receptors appear to be similar. TRAIL is a TNF-␣ homologue that has been implicated in killing by immature natural killer cells (26) as well as interferon-stimulated monocytes and dendritic cells (27,28). Apoptosis induced by this cytokine involves ligation of death receptor 4 and/or death receptor 5 followed by recruitment of FADD and caspase-8 into a DISC (29 -31). While binding of TNF-␣ to type I TNF-␣ receptor first leads to recruitment of the type I TNF-␣ receptor-associated death domain protein TRADD (32), TRADD then serves as a platform to recruit other polypeptides, including FADD, into the complex (33,34).
A number of previous studies have reported that PMA suppresses Fas-mediated apoptosis (35)(36)(37)(38)(39)(40)(41)(42)(43). Because PMA activates members of the PKC family, a group of 12 related lipiddependent serine/threonine kinases that play fundamental roles in signal transduction pathways that regulate growth and differentiation (44,45), it has been assumed that one or more PKC isoenzymes play a role in modulating the Fas pathway. How PKC activation alters Fas-mediated apoptosis, however, has remained unclear. Various studies have implicated PKC-induced activation of phosphatidylinositol-3 kinase (46) and the Raf 3 MEK1 3 ERK pathways (35,36,47,48) as well as p90 rsk -induced phosphorylation of BAD (49) in the inhibition of Fas-mediated signaling. On the other hand, PKC␣-mediated phosphorylation of Bcl-2 on Ser 70 reportedly increases the antiapoptotic function of Bcl-2 (50, 51), providing a potential explanation for the ability of PMA to block Fas-mediated apoptosis only in type II cells (42).
In the present study, we examined the effect of PMA on apoptotic events in prototypic type I and type II cells. These experiments demonstrated that PMA inhibits induction of apoptosis by FasL, TNF-␣, and TRAIL, but not ionizing radiation or topoisomerase poisons, in type II cells. Additional experiments revealed that PMA inhibits the recruitment of FADD to the DISC in type II but not type I cells. These results provide new insight into the manner in which PKC regulates death receptor signaling in type II cells.

EXPERIMENTAL PROCEDURES
Materials-Reagents were purchased from the following suppliers: PMA, camptothecin, and Hoechst 33258 from Sigma; etoposide from Biomol (Plymouth Meeting, PA); enhanced chemiluminescence reagents, Sepharose-coupled glutathione, protein A, and protein G from Amersham Biosciences, Inc.; CH.11 IgM monoclonal anti-Fas from Kamiya (Seattle, WA); TNF-␣ and TRAIL from R & D Systems (Minneapolis, MN); apo-1-1 IgG 1 monoclonal anti-Fas, BisVIII, H7, calphostin C, and chelerythrin from Alexis (San Diego, CA); monoclonal antibodies to procaspase-8 and procaspase-3 from Pharmingen (La Jolla, CA); murine monoclonal E10 that recognizes ERK phosphorylated on Thr 202 and Tyr 204 as well as rabbit antiserum that recognizes modified and unmodified ERK from Cell Signaling Technology (Beverly, MA); rabbit polyclonal anti-mouse IgM from Dako; and horseradish peroxidaseconjugated anti-mouse IgG, IgG 1 , and IgG 2a from Southern Biotechnology (Birmingham, AL). Murine monoclonal antibodies that recognize poly(ADP-ribose) polymerase and heat shock protein 90 were gifts from Dr. G. Poirier (Laval University, Ste-Foy, Quebec, Canada) and David Toft (Mayo Clinic, Rochester, MN), respectively. The rabbit antiserum that recognizes a neoepitope at the C terminus of the caspase-3 large subunit was previously characterized (52).
Induction and Assessment of Apoptosis-The human T lymphoblastoid cell line Jurkat and B lymphoma cell line SKW6.4 from the Amer-ican Type Culture Collection (Manassas, VA) were maintained at concentrations of Ͻ1 ϫ 10 6 cells/ml in RPMI 1640 medium containing 5% heat-inactivated fetal bovine serum, 100 units/ml penicillin G, 100 g/ml streptomycin, and 2 mM glutamine. To start the experiments, cells were treated with 50 nM PMA (or 0.1% dimethyl sulfoxide) for 3-5 min prior to the addition of a variety of death stimuli. After incubation as indicated in the figures, cells were sedimented at 200 ϫ g for 5 min and washed with ice-cold PBS. Two methods were then used to assess apoptosis. For flow cytometric analysis, cells were fixed in 50% ethanol for 30 min, washed with PBS, treated for Ն30 min at 20°C with 10 g/ml RNase A and 20 g/ml propidium iodide, and subjected to flow cytometry on a Becton-Dickinson FACScan using an excitation wavelength of 488 nm and an emission wavelength of 617 nm. Sample acquisition and analysis were performed with Becton-Dickinson CellQuest software. Cells containing Ͻ2 N DNA were considered apoptotic and expressed as a percentage of the total cells (53). Alternatively, cells were fixed in 3:1 (v/v) methanol/acetic acid, stained with 1 g/ml Hoechst 33258, and examined by fluorescence microscopy. Cells containing nuclei that were condensed or fragmented into multiple discrete fragments as illustrated in Fig. 2 were scored as apoptotic. At least 400 cells/sample were counted.
Analysis of the DISC by Immunoprecipitation-Aliquots containing 5 ϫ 10 7 cells were treated with 500 ng/ml CH.11 anti-Fas antibody for 90 min in the absence or presence of 50 nM PMA. All further steps were performed at 4°C. Cells were washed once with PBS and lysed by incubation for 20 min in 1 ml of DISC buffer. Insoluble debris was removed by centrifugation for 15 min at 14,000 ϫ g. As a control, 1 ml of cell lysates from 5 ϫ 10 7 untreated cells were supplemented with 1 g of CH.11 antibody. After protein concentration of the extracts was determined, cell lysates containing 5 mg of protein in 1 ml DISC buffer were incubated with 10 g/ml rabbit anti-mouse IgM for 2 h. Aliquots of protein A-Sepharose (30 l) and protein G-Sepharose (10 l) were added for an additional 2 h. Immune complexes were then pelleted by centrifugation for 3 min at 14,000 ϫ g, washed five times with DISC buffer, and released from the beads by boiling for 5 min in SDS sample buffer. Samples were subjected to SDS-PAGE, transferred to nitrocellulose, and sequentially probed with reagents that recognize FADD, procaspase-8, and Fas. Blots were scanned using Adobe Photoshop and quantitated using NIH Image 1.61 as previously described (55).
Procaspase-8/FADD Binding-The pGex4T-2 construct containing GST fused to full-length FADD was kindly provided by Dr. M. Peter (University of Chicago). Escherichia coli strain DH5␣ cells transformed with this construct were grown overnight and induced with 1 mM isopropyl-D-thiogalactoside for 3 h. GST-FADD was bound to glutathione-Sepharose beads according to the manufacturer's protocol and eluted by incubation with 10 mM glutathione in 50 mM Tris-HCl (pH 9.0). The eluates were dialyzed against PBS overnight. After Jurkat cells were treated with diluent or 50 nM PMA for 90 min, cell lysates were prepared by freezing and thawing cells in PBS containing the same protease and phosphatase inhibitors described for DISC buffer. Aliquots containing 5 mg of protein were incubated for 3 h at 4°C with purified GST-FADD. Glutathione-Sepharose was added for an additional 2 h. FADD-bound proteins were then pelleted by centrifugation for 3 min at 14,000 ϫ g, washed five times with PBS containing protease and phosphatase inhibitors, and released from the beads by boiling for 5 min in SDS sample buffer. Recovery of procaspase-8 was assessed by immunoblotting.

RESULTS
PMA Inhibits Fas-mediated Apoptosis in Jurkat Cells but Not SKW6.4 Cells-During Fas-mediated apoptosis, the events downstream of caspase-8 activation at the DISC are different in type I and type II cells (17,42). To confirm previous reports that PMA differentially inhibited Fas-mediated apoptosis in the two cell types, we treated prototypic type II (Jurkat) and type I (SKW6.4) cells with the agonistic anti-Fas antibody CH.11 in the absence or presence of PMA. Treatment of Jurkat cells with CH.11 resulted in the induction of apoptosis as indicated by the presence of increasing numbers of subdiploid cells after ethanol fixation followed by propidium iodide staining (Fig. 1A), by proteolytic cleavage of procaspase-8 and -3 (Fig.  1B, upper panels, lanes 6 -8), by cleavage of the caspase substrate poly(ADP-ribose) polymerase (Fig. 1B, lower panel, lanes 6 -8), and by the presence of apoptotic morphological changes after fixation and staining with Hoechst 33258 (Fig. 2A). When 50 nM PMA was added 3-5 min prior to CH.11, development of all of these signs of apoptosis was inhibited (Figs. 1, A and B,  and 2A).
Treatment of SKW6.4 cells with CH.11 also resulted in apoptosis ( Fig. 1, C and D). Compared with Jurkat cells, however, SKW6.4 cells showed little inhibition of apoptosis when PMA was added prior to CH.11 (Fig. 1C). Moreover, PMA only slightly inhibited the cleavage of procaspase-8, procaspase-3, and poly(ADP-ribose) polymerase in these cells (Fig. 1D). These results are consistent with a previous report that PMA fails to inhibit Fas-mediated apoptosis in SKW6.4 cells (41).
PMA Inhibits Death Receptor-mediated Apoptosis but Not Mitochondria-initiated Apoptosis in Jurkat Cells-The observation that PMA inhibits Fas-mediated apoptosis in Jurkat cells, which require mitochondrial amplification of the DISC- initiated proteolytic signal (17), but not in SKW6.4 cells, in which apoptosis occurs independent of mitochondrial amplification (17), raised the possibility that PMA might be acting by inhibiting the mitochondrial pathway, possibly through an effect of PKC␣ on Bcl-2 phosphorylation (50,51). To assess this possibility, we treated Jurkat cells with etoposide, a chemotherapeutic agent that induces apoptosis through a pathway that involves mitochondrial release of cytochrome c (56) and caspase-9 activation (52, 57) but is independent of Fas (58) or FADD (59) function. Interestingly, PMA did not protect Jurkat cells from etoposide-induced apoptosis (Fig. 2B). Instead, nuclear fragmentation ( Fig. 2A) and poly(ADP-ribose) polymerase cleavage (Fig. 2C) were readily detectable in cells treated with etoposide and PMA. Based on current understanding of etoposide-induced apoptosis (52, 56 -59), these results argue against the possibility that PMA is exerting its effects on Fas-mediated apoptosis by inhibiting the mitochondrial pathway.
When effects of other stimuli were evaluated, PMA also inhibited proteolytic cleavage of procaspase-8 and induction of apoptosis by FasL, TNF␣ and TRAIL in Jurkat cells (Fig. 2, B and C). In contrast, PMA did not protect from induction of apoptosis by camptothecin or ␥-irradiation. These results not only provide additional support for the view that topoisomerase poisons and ionizing radiation kill cells through a pathway that is distinct from the known death receptor pathways (13,14) but also indicate that PMA acts specifically on death receptor pathways rather than inhibiting the mitochondrial pathway.
Failure of ERK Pathway Activation to Account for the Protective Effects of PMA-Previous studies have demonstrated that PMA treatment results in activation of the MEK1 3 ERK pathway in lymphoid cells (48,60,61). To evaluate the possible role of this signal transduction pathway in the PMA-induced inhibition of CH.11-induced apoptosis, Jurkat cells were treated with PMA in the absence or presence of the MEK1 inhibitor PD98059 (62,63). Treatment with PD98059 by itself had no demonstrable effect on these cells (Fig. 3). If PMA were acting through the MEK1 3 ERK pathway, PD98059 would be expected to abolish the effects of PMA. Contrary to this prediction, PD98059 had minimal effect on the CH.11/PMA combination (Fig. 3A) despite its ability to inhibit ERK phosphorylation in these cells (Fig. 3B).
PMA Disrupts Fas Binding to FADD in Jurkat Cells but Not SKW6.4 Cells-Because the effect of PMA could not be explained by its effect on either the mitochondrial pathway of caspase activation or the mitogen-activated protein kinase pathway, we next evaluated the possibility that PMA might be altering death receptor signaling. To address this possibility, DISC components were analyzed in CH.11-treated SKW6.4 and Jurkat cells.
When SKW6.4 cells were treated with CH.11, immunoprecipitation of the antibody-bound Fas followed by immunoblotting with anti-FADD antibody revealed that the DISC formed in Ͻ5 min (Fig. 4A). Treatment with PMA had little effect on the recruitment of FADD to Fas in these cells (Fig. 5A).
Examination of Jurkat cells revealed a markedly different picture. Consistent with previous reports that DISC assembly in type II cells is difficult to detect (17), preliminary experiments indicated that the Fas-FADD interaction was almost undetectable 5 min after the addition of CH.11 to Jurkat cells (Fig. 4A, lane 7). Instead, FADD could be more readily precipitated with Fas 30 -90 min after the addition of CH.11 (Fig. 4A,  lanes 8 -10). Control experiments indicated that the delayed and diminished DISC formation in Jurkat cells occurred despite the presence of levels of Fas, FADD, and procaspase-8 that were similar to SKW6.4 cells (Fig. 4B).
This ability to detect DISC formation, although it was de-layed, provided the opportunity to assess the effect of PMA on Fas signaling events in these type II cells. In subsequent experiments, Fas immunoprecipitates prepared from cells treated for 90 min with diluent, CH.11, or CH.11 plus PMA were probed for the presence of FADD as well as procaspase-8. When Fas was immunoprecipitated by the addition of CH.11 to cell lysates prepared from untreated cells, FADD and procaspase-8 were undetectable in the immunoprecipitates (Fig.  5B, lane 1). When cells were treated with CH.11 for 90 min, FADD and procaspase-8 were both detected in the immunoprecipitates (Fig. 5B, lane 2). The addition of PMA prior to CH.11 resulted in a Ͼ90% decrease in the amount of Fas-associated FADD and procaspase-8 (Fig. 5B, lane 3).
To rule out the possibility that this decreased recovery reflected PMA-induced decreases in FADD or procaspase-8 expression, levels of these polypeptides were analyzed in cell lysates obtained from the same experiment. As indicated in the lower panels of Fig. 5B, PMA treatment did not affect total cellular expression of either FADD or procaspase-8. The decreased recovery of procaspase-8 in the DISC after PMA treatment (Fig. 5B, lane 3) might reflect decreased binding of procaspase-8 to FADD in addition to the decreased Fas-FADD interaction. To examine this possibility, we assessed the ability procaspase-8 in extracts from control or PMA-treated cells to bind to GST-FADD in vitro. Results of this analysis (Fig. 5C) indicated that PMA had no effect on the binding of procaspase-8 to FADD.
BisVIII Enhances Death Receptor-mediated Apoptosis without Enhancing DISC Formation-Based on the ability of PMA to inhibit death receptor-mediated apoptosis, we evaluated the possibility that inhibition of protein kinase C might enhance response to the same cytotoxic ligands. In a previous study, Zhou et al. (64) demonstrated that treatment of Jurkat cells with certain bisindoylmeleimides potentiated Fas-mediated apoptosis, whereas other protein kinase C inhibitors did not. Results shown in Fig. 6, A and B, demonstrate that BisVIII, the most potent of the bisindoylmeleimides, enhanced apoptosis after treatment of Jurkat cells with CH.11, TNF-␣, or TRAIL. In contrast, the protein kinase C inhibitor H7 failed to enhance apoptosis after any of these treatments. In further experiments (data not shown) the protein kinase C inhibitors calphostin C and chelerythrin induced apoptosis in Ͼ85% of Jurkat cells without the addition of death ligand, making it difficult to assess the effects of these particular protein kinase C inhibitors.
To determine whether the effect of BisVIII might be due to enhanced DISC formation, cells were pretreated with BisVIII or H7 for 5 min before treatment with CH.11. Examination of whole cell lysates (Fig. 6C, lower panels) demonstrated that BisVIII altered the phosphorylation status of FADD, as indicated by a decrease in the slower migrating band and increase in the faster migrating band (65). This was reflected in the composition of FADD that was recruited to the DISC (Fig. 6C,  upper panels). Nonetheless, the total amount of FADD in the DISC was not altered by treatment with BisVIII (Fig. 6, C and  D). H7 (Fig. 6, C and D) and chelerythrin (not shown), had no effect on either phosphorylation or recruitment of FADD. These observations suggest that the protein kinase C inhibitors examined do not affect DISC formation. DISCUSSION A number of previous studies have demonstrated that PMA inhibits Fas-mediated apoptosis in type II cells. The mechanism of this effect, however, has remained incompletely understood. Moreover, it has been unclear whether this effect would be observed when other death receptors are ligated. In the present study, we have demonstrated that PMA treatment of type II cells also inhibits the effects of TNF-␣ and TRAIL but not etoposide, camptothecin, or ionizing radiation. In addition, we have demonstrated that PMA inhibits recruitment of FADD and procaspase-8 to the Fas-induced DISC. These observations have potentially important implications for current understanding of the regulation of death receptor signaling.
The pathway leading from Fas ligation to caspase activation has been increasingly well characterized over the past 5 years (5,6,66,67). Initial events in this process involve binding of FADD to the death domain of Fas, followed by binding of procaspase-8 to the death effector domain of FADD. Once caspase-8 is activated, it can either directly activate procaspase-3 or cleave the cytosolic protein Bid to form a fragment that stimulates caspase activation through the mitochondrial pathway (22,25). After cells were treated with 500 ng/ml CH.11 for the indicated length of time, lysates containing 5 mg of protein were incubated with rabbit anti-mouse IgM followed by protein A-and protein G-Sepharose to immunoprecipitate Fas. Immunoprecipitates were subjected to SDS-PAGE and probed with anti-FADD antibody. B, whole cell lysates from SKW 6.4 or Jurkat cells were probed for FAS (immunoprecipitation followed by immunoblotting), FADD, procaspase-8, or heat shock protein 90 (immunoblotting). Equal amounts of protein cell lysate from the two cell lines were analyzed in each assay. These events can be regulated in a number of different ways. First, it has been suggested that the Fas receptor is preoligomerized and that mutations might affect this oligomerization state (68), although this preoligomerization has not been a universal finding (69). Second, it has been reported that Fas is a phosphoprotein whose signaling is regulated by an associated phosphatase (70). Third, the binding of FADD to procaspase-8 is inhibited by an endogenous polypeptide (Flice-like inhibitory protein) that bears death effector domains but no caspase active site (reviewed in Ref. 19). Fourth, it has been reported that caspase activation downstream of caspase-8 can be inhibited by inhibitor of apoptosis proteins (71). Finally, in type II cells, Fas-mediated apoptosis can be modulated by alterations that affect the mitochondrial pathway. For example, overexpression of Bcl-2 inhibits Fas-mediated apoptosis in type II cells (72)(73)(74)(75)(76).
Previous studies have suggested that PMA might be exerting its effects by inducing phosphorylation and inactivation of the proapoptotic Bcl-2 family member BAD (35,49), possibly through activation of phosphatidylinositol 3-kinase (46). It has also been reported that PMA induces Bcl-2 phosphorylation (51) and inhibits Bid cleavage (35,42). Whether these effects explain PMA-induced inhibition of Fas-mediated signaling has remained unclear. In the present study, we utilized several approaches to address this issue.
First, we examined the effect of PMA on other treatments. PMA inhibited the initiation of apoptosis by FasL, TRAIL, and TNF-␣ (Fig. 2, B and C). Although an effect of PKC activation on TRAIL-induced signaling was reported after completion of the present studies (77), to our knowledge this is the first demonstration that induction of apoptosis by FasL or TNF-␣ can be inhibited by PMA. In contrast, we observed that PMA failed to inhibit apoptosis induced by etoposide or ionizing radiation. These two stimuli initiate apoptotic signaling through a process that is independent of caspase-8 (52,78) and instead appears to involve caspase-9 as an initiating protease (59,79). Overexpression of Bcl-2 inhibits apoptosis induced by these treatments (80,81). The observation that PMA fails to protect cells from these treatments (Fig. 2, A and B) argues against the possibility that the effect of PMA on Fas-mediated apoptosis reflects enhanced Bcl-2 function. These same observations also argue against a critical role for BAD phosphorylation, another method of regulating the mitochondrial pathway, in PMA-induced protection.
Second, we examined the effect of PMA on MEK-mediated events. Previous studies have suggested that activation of the MEK 3 ERK pathway protects cells from certain proapoptotic stimuli (82). Our results confirmed that PMA activated this pathway (Fig. 3B). Nonetheless, the MEK inhibitor PD98059 FIG. 6. BisVIII treatment enhances death receptor-mediated apoptosis but not DISC formation. A, effect of BisVIII and H7 on CH.11-and TNF-␣induced apoptosis. After a 5-min pretreatment with diluent, 10 M BisVIII, or 10 M H7, Jurkat cells were treated for 6 h with 12.5 ng/ml CH.11 or 10 ng/ml TNF-␣ plus 0.5 g/ml cycloheximide. At the end of the incubation, cells were fixed as illustrated in Fig. 2A and analyzed for apoptotic nuclear changes. B, effect of BisVIII and H7 on TRAIL-induced apoptosis. After a 5-min pretreatment with diluent, 10 M BisVIII, or 10 M H7, Jurkat cells were treated for 3 h with 6.25 ng/ml TRAIL and analyzed as in A. C, effect of BisVIII and H7 on DISC formation in Jurkat cells. Top, Jurkat cells (5 ϫ 10 7 ) were treated with diluent (lanes 1 and 4), 10 M BisVIII (lanes 2 and 5) or 10 M H7 (lanes 3 and 6). Beginning 5 min later, half of the samples were treated with 500 ng/ml CH.11 for 30 min (lanes 4 -6). After cells were lysed, CH.11 was added to samples in lanes 1-3. Fas receptors with associated DISC components were immunoprecipitated as described in Fig. 4A and sequentially probed with anti-FADD and anti-Fas. Bottom panels, whole cell lysates from the same cells were probed with anti-FADD or, as a loading control, heat shock protein 90 antibody. D, the total strength of the anti-FADD signal (top and bottom bands) in immunoprecipitate lanes 4 -6 of C was quantitated as described under "Experimental Procedures" and expressed in arbitrary units.
diminished PMA-induced ERK activation without altering the effect of PMA on Fas-mediated apoptosis (Fig. 3A), suggesting that signaling through ERK is unlikely to account for the inhibitory effect of PMA on CH.11-induced apoptosis.
Finally, we examined the effect of PMA on DISC assembly. The observation that PMA inhibited recruitment of both FADD and procaspase-8 to the DISC in Jurkat cells (Fig. 5B) provides an explanation for the PMA-induced inhibition of procaspase-8 cleavage (Fig. 2B), the previously reported PMA-associated inhibition of Bid cleavage in these cells (35,42), and the PMAinduced inhibition of Fas-mediated apoptosis in these cells (Figs. 1 and 2). Further analysis indicated that the potential ability of procaspase-8 to interact with FADD in Jurkat cells was unaltered (Fig. 5C), suggesting that the primary effect of PMA was to inhibit the Fas-FADD interaction in these type II cells.
When similar experiments were performed in SKW6.4 cells, a prototypic type I cell line (17), a different picture emerged. We observed that DISC formation was much more rapid and extensive in SKW6.4 cells (Fig. 4A), raising the possibility that the distinction between type I and type II cells might reflect differences in the rate of trafficking of FADD to Fas. Moreover, we observed that PMA had no effect on FADD recruitment in SKW 6.4 cells (Fig. 5A), explaining why PMA has little effect on Fas-induced apoptosis in this cell line (Fig. 1C).
Based upon a previous observation that BisVIII, which can inhibit certain protein kinase C isoforms, enhances Fas-mediated apoptosis (64), we evaluated the possibility that protein kinase C inhibitors might enhance DISC formation in Jurkat cells. The present analysis demonstrated that BisVIII enhanced the induction of apoptosis by TNF-␣ and TRAIL as well as CH.11 (Fig. 6, A and B). In contrast, H7 had no effect on induction of apoptosis by these death receptor ligands. These observations confirm the results of Zhou et al. (64) and extend them to additional death receptor ligands. In further experiments, we examined the effect of BisVIII and H7 on DISC formation. Neither of these agents altered the amount of FADD recruited to the DISC in the presence of agonistic Fas antibody (Fig. 6, C and D). Interestingly, BisVIII appeared to alter the phosphorylation state of FADD (Fig. 6C). Given the fact that FADD phosphorylation appears to be mediated by a cell cycleregulated kinase (65), this observation suggests that BisVIII might be altering the activity of kinases other than protein kinase C. While we cannot rule out the possibility that other protein kinase C inhibitors, particularly those selective for the isoform responsible for the protective effects of PMA, might alter DISC formation, these observations rule out the possibility that protein kinase C inhibitors in general are capable of altering proximal events during Fas-mediated signaling.
Collectively, our observations have several potentially important implications. First, they confirm a recent report that protein kinase C can modulate TRAIL-induced apoptosis. Further experiments in our laboratory demonstrated that PMA also inhibits TRAIL-induced apoptosis in two other type II cell lines, HL-60 human leukemia cells and HCT116 colon cancer cells, 2 providing support for the view that events described in the present report are not unique to Jurkat cells. Second, the results described above indicate that protein kinase C activation simultaneously modulates sensitivity to multiple related cytokines, including FasL and TNF-␣, providing additional insight into the mechanisms responsible for the well established ability of PMA to inhibit killing by cytotoxic T cells and natural killer cells (83,84). Third, they provide an independent line of evidence supporting the view that certain chemotherapeutic agents induce apoptosis by a pathway that is distinct from that utilized by death receptor ligands (13,14,85). Finally, these results shift the focus of studies designed to investigate the effects of PMA on death receptor signaling. Although previous studies have emphasized effects of PMA on components further downstream in apoptotic signaling pathways, our results suggest that it might be important to determine how protein kinase C activation alters the ability of DISC components to interact with each other.