Abstract
Fas ligand and TNF-related apoptosis-inducing ligand (TRAIL) induce apoptosis in many different cell types. Jurkat T cells die rapidly by apoptosis after treatment with either ligand. We have previously shown that mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK) can act as a negative regulator of apoptosis mediated by the Fas receptor. In this study we examined whether MAPK/ERK can also act as a negative regulator of apoptosis induced by TRAIL. Activated Jurkat T cells were efficiently protected from TRAIL-induced apoptosis. The protection was shown to be MAPK/ERK dependent and independent of protein synthesis. MAPK/ERK suppressed TRAIL-induced apoptosis upstream of the mitochondrial amplification loop because mitochondrial depolarization and release of cytochrome c were inhibited. Furthermore, caspase-8-mediated relocalization and activation of Bid, a proapoptotic member of the Bcl family, was also inhibited by the MAPK/ERK signaling. The protection occurred at the level of the apoptotic initiator caspase-8, as the cleavage of caspase-8 was inhibited but the assembly of the death-inducing signaling complex was unaffected. Both TRAIL and Fas ligand have been suggested to regulate the clonal size and persistence of different T cell populations. Our previous results indicate that MAPK/ERK protects recently activated T cells from Fas receptor-mediated apoptosis during the initial phase of an immune response before the activation-induced cell death takes place. The results of this study show clearly that MAPK/ERK also participates in the inhibition of TRAIL-induced apoptosis after T cell activation.
Apoptosis or programmed cell death is important in regulating tissue homeostasis in adult organisms and during embryonic development. In the immune system, negative T cell selection, as well as termination of clonally expanded peripheral T cell populations, is conducted by apoptosis (1). Apoptosis is often initiated by external stimulation of a death receptor (DR),4 which in turn initiates an intracellular signaling cascade, eventually leading to apoptosis. Whether the signals mediated by the activated receptors will lead to apoptosis or continued proliferation is dependent on cell type and the state of differentiation. DR responses need to be carefully regulated during different dynamic processes, such as proliferation, migration, and differentiation, to maintain an accurate size of a given cell population.
The Fas ligand (FasL) (2) and the TNF-related apoptosis-inducing ligand (TRAIL) (3, 4) are members of the TNF family. Both are able to induce rapid apoptosis in potential target cells, the sensitivity of which seems to be regulated by multiple mechanisms. Among the members of the TNF family, TRAIL shows the highest homology with the FasL. TRAIL is a 40-kDa type II transmembrane protein suggested to be involved in many biological processes, such as activation-induced death of lymphocytes (5, 6, 7, 8), T cell-mediated cytotoxicity (9, 10, 11), and maintenance of immune-privileged sites (12), all functions that have been assigned also for the Fas receptor (FasR). While the involvement of FasR in these processes is relatively well established, further investigations are required to determine the exact role of TRAIL in these functions.
TRAIL signaling is mediated and regulated by four distinct receptors: DR4/TRAIL-R1 (13), DR5/TRAIL-R2 (14), decoy receptor (DcR)1/TRAIL-R3 (15), and DcR2/TRAIL-R4 (16), of which the DR4 and DR5 contain functional death domains and are able to induce apoptosis. In contrast, DcR1 and DcR2 act as inhibitory receptors by lacking complete death domains. The elevated expression of DcRs in normally growing tissues could possibly explain why TRAIL induces apoptosis in most transformed but not in normal cells (16, 17).
The apoptotic signaling pathway induced by ligation of the TRAIL receptors (TRAIL-R) is still fairly uncharacterized. Fas-associated death domain (FADD) (18) and caspase-8 (19) have been previously established as important components in the FasR death-inducing signaling complex (DISC) (20). In this sense, both FADD and caspase-8 have also been indicated as crucial elements in the TRAIL-mediated signaling machinery (21, 22, 23). Activation of caspase-8 in the DISC results in activation of downstream caspases and cleavage of cytosolic substrates such as Bid (24). Bid engages a mitochondrial amplification pathway, which has been suggested to be required for induction of apoptosis in some cell types. The cleaved or truncated Bid (tBid) (25) translocates to the mitochondria, where it triggers depolarization of the mitochondria. In concert with the altered mitochondrial membrane potential (MMP), cytochrome c (cyt c) is released to the cytosol, where it forms the apoptosome together with apoptosis protease-activating factor 1 and caspase-9 (26). In turn, caspase-9 can activate downstream caspase-3 or boost the activation of other caspases, such as caspase-8, to complete the mitochondrial amplification loop (reviewed in Ref. 27). The relative importance of the mitochondrial amplification loop in DR-mediated apoptosis is still not fully understood, and both mitochondria-dependent and -independent activation mechanisms have been identified. The available information on the involvement of the mitochondrial activation in TRAIL-R signaling is very scarce.
Suppression of apoptosis has been shown to be of major importance during many physiological as well as pathological processes. Apoptosis can be negatively regulated by inhibitor proteins, such as Bcl family proteins (reviewed in Ref. 28), FLIPs (reviewed in Ref. 29), or inhibitors of apoptosis protein (reviewed in Ref. 30). Another mode of regulation is through expression of DcRs, whose presence has been described for both the Fas (31) and the TRAIL-R system (reviewed in Ref. 32). Finally, protein kinase-mediated signaling has been described as an effective way of directing DR signals (reviewed in Ref. 33). In contrast to regulation by inhibitor proteins and DcRs, phosphorylation-based signaling occurs without requirement of newly synthesized proteins. In this respect, especially the classical mitogen-activated protein kinase (MAPK) signaling pathway has been implicated as a dominant negative regulator of DR-mediated apoptosis. We have observed that MAPK/extracellular signal-regulated kinase (ERK) signaling potently modifies FasR responses (34, 35) and found indications that it is involved in regulating also TRAIL-R responses (36). We have also shown that MAPK/ERK signaling from the TCR is able to protect T cells from FasR-mediated apoptosis (37) before they commit activation-induced cell death (AICD). Therefore, we wanted to test whether this type of regulation could also apply for the TRAIL-Rs. The results of the present study show that the MAPK/ERK pathway in activated Jurkat T cells suppresses TRAIL-mediated apoptosis in a similar fashion as it suppresses FasR-mediated apoptosis. Because little was known about the role of the mitochondrial amplification loop in TRAIL-R-mediated signaling, we paid special attention to clarifying where the inhibition takes place in relation to the proapoptotic mitochondrial signaling sequence. Our results show that MAPK/ERK abrogates the apoptotic signal upstream of the mitochondrial amplification loop by inhibiting initiator caspase activity. This mechanism could especially be involved in regulation of the persistence of peripheral T cell populations.
Materials and Methods
Cell culture
The human leukemic T cell line Jurkat (clone E6-1) was received from American Type Culture Collection (Manassas, VA). The cells were cultured in RPMI 1640 medium supplemented with 10% FCS, 2 mM l-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin at 37°C in 5% CO2 in air. The cells were kept at a density of 0.5–1.0 × 106/ml.
Jurkat T cells were incubated at a density of 1 × 106/ml with TRAIL (100 ng/ml; Alexis, Läuflingen, Switzerland) along with 2 μg/ml cross-linking FLAG-tagged Ab M2 (Sigma-Aldrich, St. Louis, MO) or 100 ng/ml agonistic anti-human FasR IgM Ab (MBL, Watertown, MA) for the indicated time periods in the absence or presence of 100 μg/ml immobilized OKT3 (R.W. Johnson Pharmaceutical Institute, Bassersdorf, Switzerland) as described earlier (37), 20 nM tetradecanoyl phorbol acetate (TPA; Sigma-Aldrich), 30 μM PD 98059 (Calbiochem, La Jolla, CA), or 5 μM cycloheximide (CHX; Sigma-Aldrich).
Analysis of phosphatidylserine exposure
To detect phosphatidylserine exposure by flow cytometry, Jurkat T cells were washed once with PBS and incubated for 10 min in 400 μl binding buffer (2.5 mM HEPES/NaOH (pH 7.4), 35 mM NaCl, 0.625 mM CaCl2) with 1 μl annexin V-FITC (Alexis) and analyzed on a FACScan flow cytometer (BD Biosciences, San Jose, CA) or viewed under a RMB epifluorescence microscope (Leica, Deerfield, IL).
Immunoblotting techniques used
Immunoblotting was performed by lysing cells in Laemmli sample buffer and then resolving the proteins on a 12.5% SDS-PAGE. The separated proteins were transferred to a nitrocellulose membrane (Schleicher & Schuell, Dassel, Germany) probed with the specific Ab to ERK2 (BD Transduction Laboratories, Lexington, KY), phospho-ERK1/2 (New England Biolabs, Boston, MA), caspase-8 (a kind gift from P. Krammer, German Cancer Research Center, Heidelberg, Germany) (38), Bid (Santa Cruz Biotechnology, Santa Cruz, CA), Hsc70 (StressGen Biotechnologies, Victoria, British Columbia, Canada), or actin (Sigma-Aldrich), followed by coupling to the appropriate HRP-conjugated secondary Abs and visualization with the ECL system (Amersham, Little Chalfont, U.K.)
Transfection studies
Cells were transiently transfected by electroporation (220 V, 975 μF) in 400 μl of OptiMeM (Life Technologies, Rockville, MD) and allowed to rest for 48 h before treatments. The DNA constructs used were pMCL-HA-MKK1-K97 M and pMCL-HA-MKK1-S218E/S222, encoding for hemagglutinin (HA)-tagged dominant negative and constitutively active forms of MAPK kinase 1 (MKK1). The plasmid was a kind gift from N. Ahn (University of Colorado, Boulder, CO). Mock transfections were conducted using a pIRES-EGFP plasmid (Clontech Laboratories, Palo Alto, CA). For detection of transfected cells, the cells were fixed with 3% formaldehyde in PBS and permeabilized with 0.1% Triton X-100 (Sigma-Aldrich). After washing and blocking, cells were incubated with 10 μg/ml of a monoclonal HA-specific Ab (12CA5; Boehringer Mannheim, Mannheim, Germany) followed by incubation with FITC-conjugated anti-mouse secondary Ab and 10 mg/ml Hoechst 33342 (Molecular Probes, Eugene, OR). Bid-GFP transfection studies were conducted as mentioned above. The plasmid was a kind gift from G. J. Gores (Mayo Clinic, Rochester, MN) (39). For detection of apoptotic nuclei, cells were labeled with Hoechst 33342. Cells were finally mounted in 50% glycerol and viewed under a Leica RMB epifluorescence microscope.
Measurement of MMP by confocal microscope
To measure MMP, Jurkat T cells were equilibrated with 50 nM tetramethyl rhodamine methyl ester (TMRM; Molecular Probes) in RPMI 1640 medium supplemented with 25 mM HEPES (pH 7.2) for 1 h at 37°C in the dark. Subsequently, TRAIL was added to the equilibration medium. Leica TCS SP confocal microscope with 63× NA 1.4 oil immersion planapochromat objective was used to collect TMRM and transmission images at given time points. Red fluorescence of TMRM was imaged by using 568 nm excitation light from argon/krypton laser and emitted light was collected through 575–705 nm.
cyt c immunofluorescence analysis by confocal microscope
For immunofluorescence analysis, Jurkat cells were centrifuged onto glass coverslips, washed with PBS, and fixed with 3% paraformaldehyde. Subsequently, cells were permeabilized with 0.5% Triton X-100/PBS for 10 min at room temperature. After blocking with normal goat serum (GS), samples were incubated with mouse anti-cyt c, (clone 6H2.B4, 1:150 in PBS/0.01% Triton X-100 with 1.5% GS; BD PharMingen, San Diego, CA) for 2 h in a humidified dark chamber at 37°C. After three washes with PBS/0.01% Triton X-100, samples were incubated with Alexa 488-conjugated goat anti-mouse IgG (1:150 in PBS/0.01% Triton X-100 with 1.5% GS; Molecular Probes) for 45 min in a dark chamber. After three washes with PBS/0.01% Triton X-100, nuclei of the cells were counterstained with 0.1 μg/ml 4′,6′-diamidino-2-phenylindile hydrochloride (DAPI) and coverslips were mounted on microscope slides in 80% glycerol in PBS. cyt c release and nuclear morphology of the cells were imaged by Leica TCS SP MP confocal microscope with 63× NA 1.4 oil immersion planapochromat objective. Alexa 488 fluorescence was excited by using a 488-nm excitation line from argon/krypton laser and emission window was set at 492–560 nm. DAPI fluorescence was imaged by using a 780-nm excitation light from Ti-Sapphire (Tsunami; Spectra Physics, Mountain View, CA) laser and emission light was recorded through 400–490 nm.
Surface expression analysis of DR4 and DR5
A total of 0.5 × 106 cells were treated with TPA for the indicated time points with or without TRAIL. After washing, cells were blocked for 30 min with 1% BSA in PBS. Cells were then incubated with 1 μg of Abs to DR4 or DR5 (Alexis) in 1% BSA in PBS for 30 min followed by washing with PBS. Finally, cells were incubated with Alexa 488-conjugated goat anti-mouse IgG (Molecular Probes) for 30 min. After washes cells were analyzed on a FACScan flow cytometer. Only secondary Ab was used as a control.
TRAIL-R immunoprecipitation and DISC analysis
A total of 2 × 108 Jurkat cells per sample were left untreated or pretreated with 20 nM TPA in a 37°C water bath at cell densities between 1 and 2 × 106/ml. After 15 min cells were pelleted at 500 × g for 7 min and resuspended in 1 ml prewarmed RPMI medium. To stimulate TRAIL-Rs, 1 μg FLAG-tagged recombinant human soluble TRAIL (Alexis) and 2 μg anti-FLAG monoclonal M2 Ab (Sigma-Aldrich) were added to the cell suspension. Cells were incubated in a 37°C water bath for 15 min and the reaction was stopped by adding 10 ml of ice-cold PBS to the cell suspension. Cells were pelleted, washed with ice-cold PBS, and lysed in 1 ml lysis buffer (20 mM Tris-HCl (pH 7.4), 150 mM NaCl, 10% glycerol, 0.2% Nonidet P-40, 0.1% deoxycholate, and complete protease inhibitor mixture (Roche, Basel, Switzerland) for 30 min on ice. The cell debris was removed by centrifugation at 15,000 × g for 15 min at 4°C. The amount of protein was determined by Bradford assay and an equal amount of protein from each sample was precleared with 50 μl of Sepharose-CL-4B for 2 h at 4°C. A total of 5 μg of monoclonal anti-DR5 and 2.5 μg monoclonal anti-DR4 (Alexis) were added to samples and immunoprecipitated with 15 μl protein G beads (Amersham) for 2.5 h at 4°C. Beads were washed six times in 1 ml lysis buffer, resuspended in 3× Laemmli sample buffer, and boiled for 3 min. About one-third of immunoprecipitation samples and 20–50 μg protein from cell lysates were analyzed by 12.5 or 10% SDS-PAGE. Western blot was performed with anti-DR5 (Alexis), anti-FADD (BD Transduction Laboratories), caspase-8 (C15 caspase-8 Ab, a kind gift from P. Krammer, German Cancer Research Center), and anti-FLIP (Alexis) as described above.
Results
TPA and OKT3 suppress TRAIL-induced apoptosis through a MAPK/ERK-dependent mechanism
To study whether MAPK/ERK activation is able to modulate TRAIL-induced apoptosis of Jurkat T cells, we pretreated cells with two known MAPK/ERK activators (34, 37), the phorbol ester TPA or OKT3, the latter of which is an agonistic Ab to CD3 of the TCR complex. Our results show that pretreatment with both TPA and OKT3 suppresses TRAIL-induced apoptosis (Fig. 1, A–C). Apoptosis was measured by flow cytometric analysis of phosphatidylserine exposure on the cell membrane with annexin V conjugated to FITC (Fig. 1, B and C). Incubation with TRAIL alone induced rapid apoptosis in the cells. After 2 h almost 50% of the cells were apoptotic, whereas pretreatment with TPA or OKT3 efficiently suppressed TRAIL-induced apoptosis, as indicated by decreased phosphatidylserine exposure (Fig. 1, A and B) and DNA fragmentation (data not shown). Furthermore, the T cell activator OKT3 was able to suppress TRAIL-induced apoptosis for at least 12 h (Fig. 1 D). These results on inhibition of TRAIL-induced apoptosis by MAPK/ERK activators correspond well to our previous results showing that MAPK/ERK signaling is an effective inhibitor of FasR-mediated apoptosis (34), as well as our data showing that activated T cells stay insensitive to FasR-mediated apoptosis as long as their MAPK/ERK activity is elevated (37).
TRAIL-induced apoptosis is inhibited by TPA and OKT3 through a MAPK/ERK-dependent mechanism that is independent of protein synthesis. A, Jurkat T cells were stimulated with or without TPA (20 nM), TRAIL (100 ng/ml and 2 μg/ml oligomerizing Ab M2), or TPA and TRAIL. After 2 h the cells were labeled with annexin V-FITC and Hoechst 33342 for detection of apoptotic cells with fluorescence microscopy. Arrows indicate the apoptotic cells. B, The histograms show quantification of the number of apoptotic cells after the indicated treatments. Cells were labeled with annexin V-FITC and analyzed using a FACScan flow cytometer. Percentages of apoptotic cells with phosphatidylserine exposure are shown. C, Jurkat T cells were preincubated with 20 nM TPA (10 min) or immobilized OKT3 (30 min) before addition of 100 ng/ml TRAIL and 2 μg/ml of an oligomerizing Ab (M2). After 2 h the proportion of apoptotic cells was determined by FACS analysis of annexin V-FITC-labeled cells. Bars indicate the percentage of cells with exposed phosphatidylserine. Pretreatment of cells with 30 μM PD 98059 (30 min) before incubation with TPA, OKT3, and TRAIL abrogated the protective effect of TPA and OKT3. To study whether the observed protection from TRAIL-induced apoptosis is protein synthesis independent, we pretreated Jurkat T cells for 15 min with 5 μM CHX before the addition of TPA and/or TRAIL. D, OKT3 suppressed TRAIL-induced apoptosis for 12 h. Cells were incubated with immobilized OKT3 for the indicated time points and then stimulated with TRAIL for 2 h and analyzed as in C. The data represent mean values (mean ± SEM) from a minimum of three separate experiments.
TRAIL-induced apoptosis is inhibited by TPA and OKT3 through a MAPK/ERK-dependent mechanism that is independent of protein synthesis. A, Jurkat T cells were stimulated with or without TPA (20 nM), TRAIL (100 ng/ml and 2 μg/ml oligomerizing Ab M2), or TPA and TRAIL. After 2 h the cells were labeled with annexin V-FITC and Hoechst 33342 for detection of apoptotic cells with fluorescence microscopy. Arrows indicate the apoptotic cells. B, The histograms show quantification of the number of apoptotic cells after the indicated treatments. Cells were labeled with annexin V-FITC and analyzed using a FACScan flow cytometer. Percentages of apoptotic cells with phosphatidylserine exposure are shown. C, Jurkat T cells were preincubated with 20 nM TPA (10 min) or immobilized OKT3 (30 min) before addition of 100 ng/ml TRAIL and 2 μg/ml of an oligomerizing Ab (M2). After 2 h the proportion of apoptotic cells was determined by FACS analysis of annexin V-FITC-labeled cells. Bars indicate the percentage of cells with exposed phosphatidylserine. Pretreatment of cells with 30 μM PD 98059 (30 min) before incubation with TPA, OKT3, and TRAIL abrogated the protective effect of TPA and OKT3. To study whether the observed protection from TRAIL-induced apoptosis is protein synthesis independent, we pretreated Jurkat T cells for 15 min with 5 μM CHX before the addition of TPA and/or TRAIL. D, OKT3 suppressed TRAIL-induced apoptosis for 12 h. Cells were incubated with immobilized OKT3 for the indicated time points and then stimulated with TRAIL for 2 h and analyzed as in C. The data represent mean values (mean ± SEM) from a minimum of three separate experiments.
To corroborate the assumption that MAPK/ERK activity could be involved in the observed suppression of TRAIL-induced apoptosis, we tested whether pretreatment with the specific MKK1 inhibitor, PD 98059, could abolish the protective effect of OKT3 and TPA. Our results clearly show that pretreatment with PD 98059 abolished the protective effect of both TPA and OKT3 (Fig. 1,C), thus indicating that the protective effect of these compounds is MAPK/ERK dependent. Also, in agreement with previous observations on the FasR (34), the protective effect was protein synthesis independent as the MAPK/ERK-mediated protection was equally efficient in the presence of CHX (Fig. 1,C). To verify that OKT3 and TPA stimulation induces phosphorylation of MAPK/ERK, we analyzed the activation of MAPK/ERK by immunoblotting with an Ab that recognizes phosphorylated MAPK/ERK (Fig. 2). Incubation with TPA and OKT3 induced an increase in MAPK/ERK activity, which was inhibited by pretreatment with PD 98059 (Fig. 2), correlating well with the relative inhibition of apoptosis observed in Fig. 1 C. The results further support that the protective effect of TPA and OKT3 is MAPK/ERK dependent.
TPA and OKT3 induce MAPK/ERK activation in Jurkat T cells. Treatment of cells with TPA (10 min) and OKT3 (30 min) induces increased MAPK/ERK phosphorylation, which is inhibited by pretreatment with PD 98059 as shown by immunoblotting with a phosphospecific MAPK Ab. The lower panel shows equal loading of ERK in all samples. A representative immunoblot from three experiments is shown.
TPA and OKT3 induce MAPK/ERK activation in Jurkat T cells. Treatment of cells with TPA (10 min) and OKT3 (30 min) induces increased MAPK/ERK phosphorylation, which is inhibited by pretreatment with PD 98059 as shown by immunoblotting with a phosphospecific MAPK Ab. The lower panel shows equal loading of ERK in all samples. A representative immunoblot from three experiments is shown.
Constitutively active MKK1 protects Jurkat T cells from TRAIL-induced apoptosis
The suppressive role of MAPK/ERK in TRAIL-induced apoptosis was verified by transient transfections with HA-tagged constitutively active and dominant negative mutants of MKK1 before treatment with TRAIL. The transfected cells were then visualized by immunofluorescence labeling of HA and by DNA labeling with Hoechst 33342 to identify the nuclear morphology of the cells (Fig. 3). Constitutively active MKK1 (MKK1-CA) rendered the cells insensitive to apoptosis induced by TRAIL, while this effect was lost when cells were transfected with the dominant negative mutant. Mock transfections with green fluorescent protein did not affect the number of apoptotic cells.
MKK1-CA protects Jurkat T cells from TRAIL-induced apoptosis. A, Representative immunofluorescence micrographs of cells transiently transfected with MKK1-CA and MKK1-DN and then treated with or without TRAIL for 2 h are shown. Nuclear alterations were visualized by Hoechst 33342 labeling and transfected cells by immunofluorescence labeling of HA. Mock-transfected cells were treated as indicated above with or without TRAIL stimulation. Arrows indicate the transfected cells. B, The percentage of apoptotic cells was counted among transfected and untransfected cells. The data represent mean values (mean ± SEM) from a minimum of three separate transfections.
MKK1-CA protects Jurkat T cells from TRAIL-induced apoptosis. A, Representative immunofluorescence micrographs of cells transiently transfected with MKK1-CA and MKK1-DN and then treated with or without TRAIL for 2 h are shown. Nuclear alterations were visualized by Hoechst 33342 labeling and transfected cells by immunofluorescence labeling of HA. Mock-transfected cells were treated as indicated above with or without TRAIL stimulation. Arrows indicate the transfected cells. B, The percentage of apoptotic cells was counted among transfected and untransfected cells. The data represent mean values (mean ± SEM) from a minimum of three separate transfections.
Activation of MAPK/ERK does not affect the levels of DR4 and DR5 on the cell surface
Sensitization to DR-mediated apoptosis could be modulated by altered surface expression of the receptors. To rule out the possibility of surface receptor down-regulation, we analyzed the relative amount of DR4 and DR5 on the surface of Jurkat T cells. Jurkat T cells were immunofluorescence labeled with mAbs to the two respective receptors and analyzed by flow cytometry. The results show that predominantly DR5 is expressed on Jurkat T cells and that MAPK/ERK activation does not affect the relative number of receptors on the cell surface after treatment with TPA for up to 2 h (Fig. 4). The amount of DR4 and DR5 on the cell surface did not change in the presence of TRAIL.
The protective effect of MAPK/ERK activation is not caused by cell surface death receptor down-regulation. The surface expression of the TRAIL-Rs DR4 and DR5 were not altered during activation of Jurkat T cells. Cells were treated with TPA alone for the indicated time points or in the presence of TRAIL for 2 h. Cells were labeled with mAbs to DR4 or DR5 and analyzed on a FACScan flow cytometer. Secondary Ab was used as a control (dashed line).
The protective effect of MAPK/ERK activation is not caused by cell surface death receptor down-regulation. The surface expression of the TRAIL-Rs DR4 and DR5 were not altered during activation of Jurkat T cells. Cells were treated with TPA alone for the indicated time points or in the presence of TRAIL for 2 h. Cells were labeled with mAbs to DR4 or DR5 and analyzed on a FACScan flow cytometer. Secondary Ab was used as a control (dashed line).
MAPK/ERK signaling inhibits release of cyt c, loss of MMP, and translocation of tBid
To study at what level in the apoptotic activation machinery the inhibitory effect of MAPK/ERK is targeted, we started by analyzing whether the mitochondrial amplification loop is affected. To assess this question, we analyzed changes in the MMP after treatment with TRAIL alone or after pretreatment with TPA. While the cells that were treated with TRAIL alone lost their MMP, cells pretreated with TPA were not affected at this level (Fig. 5,A). Also, in FasR-mediated apoptosis the MMP was lost (data not shown). The FasR-mediated decrease in MMP was also abolished by the TPA-mediated activation of MAPK/ERK (data not shown). To show that this was a MAPK/ERK-dependent effect, we pretreated the cells with the MKK1 inhibitor PD 98059. Inhibition of MAPK/ERK reinduced the loss in MMP and accelerated apoptosis. Because decreased MMP has been suggested to cause release of cyt c to the cytosol (40), it was to be expected that cyt c release would also be affected by activation of MAPK/ERK. Jurkat T cells treated with TRAIL alone or pretreated with TPA were immunolabeled for cyt c and viewed under a microscope. In control cells cyt c was located in the mitochondria, which can be seen as clusters next to the nucleus in Fig. 5 B. In apoptotic Jurkat T cells, treated with TRAIL alone for 2 h, TRAIL-caused release of cyt c from the mitochondria to the cytosol was clearly visible. When cells were pretreated with the MAPK/ERK activator TPA, the release of cyt c was inhibited. Also in this case, treatment with PD 98059 reversed this inhibition and allowed the cells to undergo apoptosis. Also, FasR-mediated cyt c release was inhibited (data not shown). Treatment with TPA or PD alone did not affect the release of cyt c.
A, MAPK/ERK activation inhibits depolarization of mitochondria. Shown are confocal laser scanning micrographs of Jurkat T cells labeled with MMP dye TMRM and then treated as indicated. After 2 h live cells were analyzed by confocal microscopy. Apoptotic cells were identified by morphology from transmission images (arrowheads). Note that the apoptotic cells have lost their MMP as determined by decreased TMRM fluorescence intensity. B, Activation of MAPK/ERK inhibits the release of cyt c. Shown are confocal laser scanning micrographs of Jurkat T cells incubated for 2 h as indicated. Cells were labeled for cyt c (signal shown in green) and DNA was labeled with DAPI (signal shown in red). Note that the apoptotic cells (arrows) show cytoplasmic localization of cyt c compared with the clustered mitochondrial localization in the control and TPA-treated cells. C, MAPK/ERK activation prevents tBid translocation to mitochondria. Jurkat T cells were transfected transiently with GFP-tagged Bid. Cells were then treated as indicated and analyzed by fluorescence microscopy to detect the localization of Bid-GFP. Activation of MAPK/ERK by TPA prevented Bid-GFP translocation to the mitochondria and apoptosis. Arrows point out the apoptotic Bid-GFP transfected cells.
A, MAPK/ERK activation inhibits depolarization of mitochondria. Shown are confocal laser scanning micrographs of Jurkat T cells labeled with MMP dye TMRM and then treated as indicated. After 2 h live cells were analyzed by confocal microscopy. Apoptotic cells were identified by morphology from transmission images (arrowheads). Note that the apoptotic cells have lost their MMP as determined by decreased TMRM fluorescence intensity. B, Activation of MAPK/ERK inhibits the release of cyt c. Shown are confocal laser scanning micrographs of Jurkat T cells incubated for 2 h as indicated. Cells were labeled for cyt c (signal shown in green) and DNA was labeled with DAPI (signal shown in red). Note that the apoptotic cells (arrows) show cytoplasmic localization of cyt c compared with the clustered mitochondrial localization in the control and TPA-treated cells. C, MAPK/ERK activation prevents tBid translocation to mitochondria. Jurkat T cells were transfected transiently with GFP-tagged Bid. Cells were then treated as indicated and analyzed by fluorescence microscopy to detect the localization of Bid-GFP. Activation of MAPK/ERK by TPA prevented Bid-GFP translocation to the mitochondria and apoptosis. Arrows point out the apoptotic Bid-GFP transfected cells.
Decrease in MMP and release of cyt c from the mitochondria have been shown to be initiated by the translocation of a 15-kDa tBid from the cytosol to the mitochondria (reviewed in Ref. 41). To address the question of whether MAPK/ERK activation affects translocation of tBid, we transiently transfected Jurkat T cells with Bid-GFP. Bid-GFP Jurkat T cells were then treated with TRAIL. In control cells, Bid-GFP showed a diffuse cytoplasmic distribution (Fig. 5,C). After 2 h in the presence of TRAIL all Bid-GFP in apoptotic cells was located in clusters near the nucleus that could be referred to as mitochondria. When the transfected cells were pretreated with TPA, Bid-GFP was no longer detected in clusters and apoptosis was inhibited. Instead, Bid was diffusely located in the cytosol similarly to control cells. Pretreatment with PD 98059 abolished the inhibition of Bid-GFP translocation (Fig. 5,C) as well as subsequent apoptosis (Fig. 1 C). Again, PD or TPA had no effect alone.
Taken together these results show that TRAIL-induced apoptosis is directed toward the mitochondrial amplification loop and that MAPK/ERK signaling protects Jurkat T cells from apoptosis by inhibiting any dysregulation of the mitochondria by turning off the mitochondrial amplification loop.
MAPK/ERK activation suppresses the cleavage of Bid and caspase-8
To examine whether the protective effect of MAPK/ERK signaling would occur at the level of the DISC, we analyzed how the cleavage of initiator caspase-8 was affected in cells treated with TPA before incubation with TRAIL. The results in Fig. 6,A show that the cleavage of caspase-8 to the active 18-kDa fragment is markedly reduced in cells pretreated with TPA. Also, processing to the intermediate 42/43-kDa fragments is reduced in the presence of TPA. Caspase-8 has previously been shown to activate Bid by cleavage to a 15-kDa tBid fragment (25). Therefore, we wanted to examine whether activation of Jurkat T cells suppresses the cleavage of Bid to its active proapoptotic 15-kDa fragment. As expected, its cleavage was reduced after preincubation of the cells with TPA (Fig. 6 B). This inhibition is likely to prevent the advancing of the apoptotic signal to the mitochondrial amplification loop, which has been implicated to be necessary for FasR-mediated apoptosis in type II cells (42).
MAPK/ERK activation suppresses cleavage of caspase-8 and Bid. Jurkat T cells were treated as indicated before assessment of caspase-8 (A) and Bid (B) cleavage by Western blotting with specific Abs. Caspase-8 activation can be observed as the appearance of an 18-kDa active fragment, whereas activation of Bid is detected by the appearance of tBid as a 15-kDa fragment. TPA treatment inhibited the cleavage of both caspase-8 and Bid, thereby preventing their activation. The lower panels show equal loading of Hsc70 and actin in all samples. Representative immunoblots from three experiments are shown.
MAPK/ERK activation suppresses cleavage of caspase-8 and Bid. Jurkat T cells were treated as indicated before assessment of caspase-8 (A) and Bid (B) cleavage by Western blotting with specific Abs. Caspase-8 activation can be observed as the appearance of an 18-kDa active fragment, whereas activation of Bid is detected by the appearance of tBid as a 15-kDa fragment. TPA treatment inhibited the cleavage of both caspase-8 and Bid, thereby preventing their activation. The lower panels show equal loading of Hsc70 and actin in all samples. Representative immunoblots from three experiments are shown.
Taken together, the results show that MAPK/ERK activation inhibits TRAIL-induced apoptosis at the early stages of the apoptotic machinery before involvement of the mitochondria.
Activation of MAPK/ERK does not affect the recruitment of FADD and caspase-8 to the TRAIL-DISC
To examine whether the observed protection mediated by MAPK/ERK activation could be located at the very early stages of DR signaling, we immunoprecipitated the TRAIL-DISC to analyze the assembly of the adapter proteins recruited to the DISC. Because Jurkat T cells have been indicated to be type II cells, only a moderate amount of DISC is formed after receptor activation, also reflected in our experiments by the relative low levels of both FADD and caspase-8 coimmunoprecipitated with DR4 and DR5 after TRAIL stimulation (Fig. 7,A). To control successful immunoprecipitation, the presence of DR5 in the immunoprecipitates was detected in the immunoprecipitated samples but not in the protein G control devoid of immunoprecipitating Abs. In our study we did not detect any changes in the amount of FADD or caspase-8 recruited to the DISC after pretreatment with TPA (Fig. 7,A). Although the overall caspase-8 cleavage is reduced by MAPK/ERK activation (Fig. 6,A), the amount of caspase-8 cleaved at the DISC is equal in the presence of TPA. Furthermore, we did not detect any changes in the recruitment of cFLIP to the DISC upon TPA stimulation (data not shown). To verify that MAPK/ERK was active during the same experiment, we also analyzed the cell lysates for phosphorylated ERK1/2. Active ERK1/2 could be detected only in the presence of TPA (Fig. 7 B). Furthermore, to verify the protecting effect of TPA under the same conditions, cells were further incubated at 37°C and later monitored for apoptosis. The cells pretreated with TPA were still protected from TRAIL-induced apoptosis after several hours (data not shown). Together these experiments show that the assembly of the TRAIL-DISC is not affected by the elevated MAPK/ERK activity as earlier shown also for the FasR-DISC (37).
MAPK/ERK activation does not affect the TRAIL-DISC. Jurkat T cells were pretreated with or without 20 nM TPA and then stimulated with TRAIL for 15 min. Then the TRAIL-DISC was immunoprecipitated with mAbs to DR4 and DR5. To rule out possible unspecific binding of proteins during immunoprecipitation, an equal amount of cell lysate was incubated with protein G beads without any immunoprecipitating Abs (prot G beads). A, The immunoprecipitates were resolved on a SDS-PAGE and analyzed for the presence of DR5, FADD, and caspase-8. Cell lysates were also analyzed for FADD and caspase-8 to show equal input of protein. Nonspecific IgG reactivity is marked with an asterisk. B, The cell lysates were also analyzed for the presence of phosphorylated MAPK/ERK (pERK) to show that MAPK/ERK was activated during the experiment. Representative immunoblots from two separate experiments are shown.
MAPK/ERK activation does not affect the TRAIL-DISC. Jurkat T cells were pretreated with or without 20 nM TPA and then stimulated with TRAIL for 15 min. Then the TRAIL-DISC was immunoprecipitated with mAbs to DR4 and DR5. To rule out possible unspecific binding of proteins during immunoprecipitation, an equal amount of cell lysate was incubated with protein G beads without any immunoprecipitating Abs (prot G beads). A, The immunoprecipitates were resolved on a SDS-PAGE and analyzed for the presence of DR5, FADD, and caspase-8. Cell lysates were also analyzed for FADD and caspase-8 to show equal input of protein. Nonspecific IgG reactivity is marked with an asterisk. B, The cell lysates were also analyzed for the presence of phosphorylated MAPK/ERK (pERK) to show that MAPK/ERK was activated during the experiment. Representative immunoblots from two separate experiments are shown.
Discussion
Although the functions and physiological roles of TRAIL-Rs are not by any means as well established as those of the FasR, it is quite obvious that the emerging view of TRAIL-R tasks converges with the established roles of the FasR/FasL system, as listed in the introduction. While many features of the TRAIL-Rs are shared by the FasR and because TRAIL shows high similarity with the FasL, it is not surprising that the physiological roles of TRAIL-Rs resemble those of FasR. In addition to the converging physiological roles, there are several recent studies to indicate also that the regulation of both receptor systems is executed by similar mechanisms. Both receptor systems mediate their signals through the assembly of a DISC (20) that holds FADD (18) and caspase-8 (19) as key apoptotic signal transducers. The signaling of both TRAIL-R and FasR are regulated by DcRs and by regulation of receptor levels present on the cell surface (31, 32), and in the cytoplasmic domains FLIP has been implicated as both a signal regulator (43, 44) and a signal conveyor (45). In addition to these shared regulatory mechanisms, all of which work at the level of protein-protein interactions, we obtained in the present study compelling evidence that MAPK/ERK is a dominant inhibiting regulatory mechanism that directs TRAIL-R signaling in the same way as FasR signaling. Similarly to the FasR, MAPK/ERK regulates TRAIL-R responses through direct posttranslational regulation, without involvement of newly synthesized proteins. This type of direct signaling-mediated regulation is likely to act in concert with the other described regulatory mechanisms and would be of special importance in dynamic situations when rapid protection or sensitization is required but there is no time to engage the transcriptional machinery or to regulate protein levels otherwise.
In T cells both FasR and FasL are up-regulated upon T cell activation and both participate later in the attenuation of immune responses, to avoid formation of autoreactive T cells (reviewed in Ref. 46). This modulation occurs through AICD, where T cells kill themselves or their neighboring cells, supposedly by activating the FasR. However, T cells are insensitive to FasR-mediated apoptosis immediately after activation, although they express both FasR and FasL. This insensitivity is important in order for the activated T cells to fulfill their task of killing the target cell. It has also been shown that TRAIL is up-regulated immediately after T cell activation (47, 48) and that TRAIL later participates in the down-regulation of immune responses. While it is still quite controversial whether TRAIL is involved in AICD of human peripheral T lymphocytes in vivo (5, 7, 8), there are results indicating that AICD of Jurkat T cells (8) as well as T cells derived from HIV patients (6) involves TRAIL. There are also indications that TRAIL would inhibit cell cycle progression, thereby arresting the T cells so they can be killed by other ligands, such as FasL (49). Therefore, it is important that recently activated T cells stay insensitive to both FasL and TRAIL during the early phase of activation so that the cells can fulfill their task of killing their target cells without being sensitive to their own death ligands.
MAPK/ERK signaling modulates apoptosis induced by TRAIL
Because our previous results indicate that the insensitivity of activated T cells to FasR-mediated apoptosis depends on MAPK/ERK activation (34, 37), we examined whether MAPK/ERK signaling could suppress TRAIL-induced apoptosis during early phases of T cell activation. Our results show that activated Jurkat T cells display similar kinetics of MAPK/ERK activation and insensitivity to TRAIL, as was shown with the FasR (37). Our results indicate that MAPK/ERK signaling mediates a protective signal to both FasR-mediated and TRAIL-induced apoptosis during these early phases of T cell activation. This protective signal is then turned off at the end of the immune response to allow AICD and attenuation of the immune response. There is one report indicating that activation of protein kinase C (PKC) can protect cells from TRAIL-induced apoptosis independently of MAPK/ERK (50). This study indicated that PKC activity is mainly responsible for the observed protection from TRAIL-induced apoptosis upon TPA treatment. Differences in the experimental setup could explain the different outcome of the experiments. It is also difficult to separate these two pathways at the level of PKC, because it is an upstream regulator of MAPK/ERK, especially by using only pharmacological signaling inhibitors and activators, the principal approach in the above-mentioned study. There is also the distinct possibility that PKC and MAPK/ERK act as separate signaling entities regulating DR responses. However, while it is plausible that PKC and possibly other signaling modulators regulate FasR and TRAIL-R sensitivity, our results undoubtedly show that the MAPK/ERK signaling pathway can function as a single dominant regulator of TRAIL responses.
The MAPK/ERK-mediated protection is independent of protein synthesis and does not alter the relative amount of DR4 or DR5 on the cell surface
It is well known that several cell lines can be sensitized to DR-mediated apoptosis by pretreatment with protein synthesis inhibitors (36). This raises the possibility that the MAPK/ERK-mediated effect could be protein synthesis dependent. However, our previous results have demonstrated that MAPK/ERK-mediated suppression of FasR-mediated apoptosis is not protein synthesis dependent (34, 35, 36). The results presented in this work show that the same is true for MAPK/ERK-mediated suppression of TRAIL-induced apoptosis. Thus, a high MAPK/ERK activity is sufficient to trigger the protective effect. Direct modulation by phosphorylation-based signaling is beneficial to quickly modulate TRAIL sensitivity in situations with rapidly fluctuating conditions, such as cell growth and differentiation. This type of rapid protein synthesis-independent regulation is likely to act in concert with various regulatory proteins (e.g., FLIP). Once such a regulatory protein has been produced, it will yield a more long-term and stable protection or modulation.
There are many possible targets for a direct signaling-mediated modulation of the DR signal. A protein synthesis-independent signal-based mechanism that regulates TNF-R1 sequestration has been reported. In this case, MAPK/ERK signaling was shown to phosphorylate TNF-R1 directly, thereby causing the internalization of the receptor from the surface of the cell to the cytosol and inhibition of its cytotoxic ability (51, 52). However, our study excludes the possibility of TRAIL-R internalization or down-regulation as a MAPK/ERK target, because the surface expression of the DR4 and DR5 was not altered by the activation of MAPK/ERK.
It is tempting to speculate that the protective action of MAPK/ERK would be mediated by direct or indirect phosphorylation of a DISC component. Phosphorylation of FADD has previously been indicated (53, 54). However, our results show that MAPK/ERK activation does not alter the binding of FADD or caspase-8 to the TRAIL-DISC. Therefore, it is unlikely that MAPK/ERK activation would affect the assembly of the DISC by phosphorylation of a DISC protein. Recent findings also show that FasR, DR4, and DR5 are not phosphorylated by active MAPK/ERK or by TPA (55). It seems that finding the potential MAPK/ERK target responsible for conveying the observed protection will require detailed phosphoprotein analysis far beyond the DISC.
MAPK/ERK activation inhibits the processing of caspase-8 and Bid, thereby turning off the mitochondrial amplification loop
To resolve whether the cells still maintain their normal functions, it is of great interest to know whether MAPK/ERK suppresses TRAIL-induced apoptosis at the same stage of the signaling pathway as it does in FasR-mediated apoptosis. Similarly to the FasR (34, 37), MAPK/ERK signaling seemed to suppress TRAIL-induced apoptosis before activation of caspase-8 and Bid. It has been shown that TRAIL induces caspase-8-mediated cleavage of Bid in other cell systems (56, 57, 58). Because MAPK/ERK signaling appeared to inhibit activation of both caspase-8 and Bid, we wanted to see whether inhibition of this upstream activator suppressed all traces of downstream mitochondrial activation. It has not yet been clear whether the mitochondrial pathway is involved in TRAIL-induced apoptosis. The results in this work show that Bid is rapidly cleaved with simultaneous translocation of tBid in TRAIL-stimulated cells. MAPK/ERK activation inhibits both cleavage and translocation of tBid from the cytoplasm to the mitochondria, as well as the observed loss in MMP and release of cyt c to the cytosol, after stimulation of the FasR and TRAIL-Rs. Normally during apoptosis, released cyt c binds to apoptosis protease-activating factor 1 in the cytosol to form the apoptosome where caspase-9 is activated. Caspase-9 accelerates the cleavage of caspase-8 by the proposed mitochondrial amplification loop suggested in type II cells (42). The results presented in this work show that MAPK/ERK signaling completely abrogates the engagement of mitochondria during TRAIL-induced apoptosis. Furthermore, our results indicate that the activation of TRAIL-induced cyt c release is truly dependent on the cleavage of Bid, because there were no traces of cyt c release when Bid cleavage was inhibited. Taken together, all of our results demonstrate that the inhibition of the apoptotic TRAIL signal occurs at the very proximal stages of apoptotic signaling. Inhibition at the site of death signal initiation would be a favorable way to abrogate the death signal, because the cell can thus avoid any partial damage and survive unaffected.
The elevated MAPK/ERK signaling after activation of T cells enters and inhibits the DR pathway, thereby allowing the cells to live long enough to fulfill their tasks. When the T cells are no longer needed, the MAPK/ERK activity and the levels of inhibitory proteins decrease as a consequence of insufficient activating signals, thereby allowing the cells to die by AICD. In future studies it will be of great importance to determine the molecular mechanisms and targets underlying the MAPK/ERK-mediated inhibition of apoptotic signaling in both TRAIL-induced and FasR-mediated apoptosis. Defining these targets will have great potential in treatments of various disorders related to the functions of these receptors.
Acknowledgements
We thank Natalie Ahn (University of Boulder, Boulder, CO) for the MKK1 plasmids, Gregory J. Gores (Mayo Clinic) for the Bid-GFP plasmid, Peter Krammer (German Cancer Research Center) for caspase-8 Ab, and the R. W. Johnson Pharmaceutical Research Institute (Bassersdorf, Switzerland) for providing the OKT3 Ab. We also thank all the members of our laboratory for critical comments on the manuscript and technical help during the course of this study.
Footnotes
This work was supported by Academy of Finland Grant 35718, the Sigrid Jusélius Foundation, the Erna and Victor Hasselblad Foundation, the Finnish Cancer Foundation, the Cell Signaling Program of Åbo Akademi University, and the Turku Graduate School of Biomedical Sciences (to M.P.).
Abbreviations used in this paper: AICD, activation-induced cell death; CHX, cycloheximide; cyt c, cytochrome c; DAPI, 4′,6′-diamidino-2-phenylindile hydrochloride; DcR, decoy receptor; DISC, death-inducing signaling complex; DR, death receptor; ERK, extracellular signal-regulated kinase; FADD, Fas-associated death domain; FasL, Fas ligand; FasR, Fas receptor; GS, goat serum; HA, hemagglutinin; MAPK, mitogen-activated protein kinase; MKK1, MAPK kinase 1; MMP, mitochondrial membrane potential; PKC, protein kinase C; tBid, truncated Bid; TPA, tetradecanoyl phorbol acetate; TMRM, tetramethyl rhodamine methyl ester; TRAIL, TNF-related apoptosis-inducing ligand; TRAIL-R, TRAIL receptor.