The Extracellular Signal-regulated Kinase Pathway Is Required for Activation-induced Cell Death of T Cells*

T cells can undergo activation-induced cell death (AICD) upon stimulation of the T cell receptor-CD3 complex. We found that the extracellular signal-regulated kinase (ERK) pathway is activated during AICD. Transient transfection of a dominant interfering mutant of mitogen-activated/extracellular signal-regulated receptor protein kinase kinase (MEK1) demonstrated that down-regulation of the ERK pathway inhibited FasL expression during AICD, whereas activation of the ERK pathway with a constitutively active MEK1 resulted in increased expression of FasL. We also found that pretreatment with the specific MEK1 inhibitor PD98059 prevented the induction of FasL expression during AICD and inhibited AICD. However, PD98059 had no effect on other apoptotic stimuli. We found only very weak ERK activity during Fas-mediated apoptosis (induced by Fas cross-linking). Furthermore, preincubation with the MEK1 inhibitor did not inhibit Fas-mediated apoptosis. Finally, we also demonstrated that pretreatment with the MEK1 inhibitor could delay and decrease the expression of the orphan nuclear steroid receptor Nur77, which has been shown to be essential for AICD. In conclusion, this study demonstrates that the ERK pathway is required for AICD of T cells and appears to regulate the induction of Nur77 and FasL expression during AICD.

AICD 1 was first described in T cell hybridomas and is defined as apoptosis of lymphocytes by any signal that results in lymphocyte activation and, in particular, by stimulation of the TCR⅐CD3 complex (or B cell receptor complex) with antigens or antibodies (1,2). AICD is thought to play an important role in the deletion of (a) autoreactive T cell clones in the thymus (negative selection), (b) autoreactive T cells in the periphery with specificity for autoantigens that are not presented in the thymus (peripheral tolerance), and (c) activated T cells at the termination of an immune response (1).
Several studies have shown that the expression and interaction of Fas and FasL, which results in autocrine stimulation of the Fas death pathway, is required for AICD of T hybridoma cells, Jurkat T leukemia cells, and activated T cells (3)(4)(5)(6). AICD could be inhibited with nonstimulatory anti-Fas (CD95) antibodies or with Fas-IgFc fusion protein. Recent studies have shown that the expression of FasL during AICD is dependent on an intact CD3chain and Lck, ZAP-70, CD45, calcineurin, and Ras activities (7)(8)(9)(10)(11). However, negative selection in the thymus was not affected in lpr or gld mice, which are deficient in Fas and FasL expression (12), suggesting that the Fas pathway is not an absolute requirement for thymic negative selection.
In T cells, stimulation of the TCR⅐CD3 complex results in activation of the Ras-Raf-MEK1/2-ERK1/2 pathway and, in combination with signals through the calcium/calcineurin pathway and the Jun N-terminal kinase pathway, leads to IL-2 production (reviewed in Ref. 13). T cell activation through the TCR⅐CD3 complex can be inhibited by dominant interfering Ras or dominant interfering Raf. ERK1/2 activity is also required for thymic positive selection in vivo, as demonstrated in mice expressing a dominant interfering form of Ras (14) or a dominant interfering form of MEK1 (15).
ERK1 and -2 are proline-directed protein serine/threonine kinases that belong to the family of MAPKs and phosphorylate Ser/Thr-Pro motifs in various substrates (reviewed in Ref. 16). The ERK pathway can be activated by many stimuli, including growth factors (such as epidermal growth factor and nerve growth factor), cytokines (such as IL-2), antigen receptor stimulation (TCR), insulin, and v-Src expression. ERK1 and -2 are directly activated through dual phosphorylation of tyrosine and threonine residues in a conserved TEY motif by the specific protein MAPK kinases MEK1 and -2. MEK1 and -2 can be activated by the MAPK kinase kinase Raf. The initiation of this Raf-MEK1/2-ERK1/2 cascade occurs through the recruitment of Raf to the plasma membrane by the guanine nucleotide binding protein p21 ras .
Nur77 is an orphan nuclear steroid receptor (17) and is essential for AICD in T cell hybridomas and thymocytes, but not for IL-2 secretion (18,19). To date, no ligand for Nur77 has been identified. Rapid Nur77 mRNA expression can be found 30 min after stimulation of T cell hybridoma cells with anti-CD3 antibody, which induces apoptotic DNA ladders as early as 4 h poststimulation (18). In contrast, no Nur77 expression was found during dexamethasone-induced apoptosis or IL-2 withdrawal apoptosis of CTLL-20 cells (18,19). A dominantinterfering truncated Nur77 or antisense Nur77 was capable of preventing AICD (but not IL-2 production) after TCR stimula-* This work was supported by National Institutes of Health Program Project Grant 5PO1 CA39542 (to S. J. B.) and CA09382 (to R. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
In this study, we define a role for the ERK pathway in AICD and demonstrate its requirement in the up-regulation of FasL and expression of Nur77.
Intracytoplasmic Fluorescence Staining-Cells (5 ϫ 10 6 /ml) were washed twice with PBS and incubated for 20 min at room temperature with 4% formaldehyde. After three washes with PBS, cells were incubated with 0.1% Triton X-100 (in PBS) for 2 min at room temperature, washed three times with PBS, and incubated with PBS, 2% bovine serum albumin, 0.1% azide for 10 min at 4°C. Subsequently, cells were incubated with anti-FasL-PE antibody for 30 min at 4°C, washed with PBS (with bovine serum albumin and azide), and analyzed with a flow cytometer.
In Vitro Kinase Assay and Western Blot Analysis-In vitro kinase assays and Western blot analyses were performed as described by Hibi et al. (24) with modifications. Briefly, after freezing at Ϫ70°C, cell pellets were lysed for 20 min on ice with 50 l of lysis buffer containing 20 mM Hepes (pH 7.5), 0.3 M NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.1% Triton X-100, 1 mM dithiothreitol, 0.5 mM vanadate, 10 mM NaF, 1 mM phenylmethylsulfonyl fluoride, and 2 g/ml leupeptin, aprotinin, and pepstatin. The lysate was cleared by high speed centrifugation (13,000 rpm for 10 min) and diluted at 1:3 with equilibration buffer consisting of 20 mM Hepes (pH 7.5), 2.5 mM MgCl 2 , 0.1 mM EDTA, 0.05% Triton X-100, 1 mM dithiothreitol, 0.5 mM vanadate, 10 mM NaF, 1 mM phenylmethylsulfonyl fluoride, and 2 g/ml leupeptin, aprotinin, and pepstatin. For in vitro kinase assays, the lysate was incubated for 3 h at 4°C with anti-ERK1 antibodies (5 l) and protein A-Sepharose beads (30 l of 50% slurry; Amersham Pharmacia Biotech). The precipitates were washed twice with wash buffer containing 20 mM Hepes (pH 7.5), 0.05 M NaCl, 2.5 mM MgCl2, 0.1 mM EDTA, 0.05% Triton X-100, and 0.1 mM vanadate; twice with high salt buffer containing 10 mM Tris (pH 7.5), 0.5 M LiCl, and 0.1 mM vanadate; twice with low salt buffer containing 10 mM Tris (pH 7.5), 0.1 M NaCl, 1 mM EDTA, and 0.1 mM vanadate; and finally resuspended in wash buffer. The precipitate was then incubated for 20 min at 30°C with 3 g of myelin basic protein and 10 Ci of [␥-32 P]ATP in 2ϫ kinase buffer containing 20 mM Hepes (pH 7.5), 20 mM MgCl 2 , 0.03 mM ATP, 2 mM dithiothreitol, and 0.2 mM vanadate. The reaction was stopped with 4ϫ Laemmli sample buffer. The reaction mixtures (for in vitro kinase assays) or lysates (for Western blot analysis) were resolved by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. The membranes were then developed by autoradiography (for in vitro kinase assays) or blocked with 5% nonfat milk powder in Tris-buffered saline, probed with primary antibody followed by horseradish peroxidase-conjugated secondary antibody and developed by ECL (Amersham Pharmacia Biotech) (for Western blot analysis).
Apoptosis Assay-For cross-linking experiments, cell culture dishes were coated overnight at 4°C with anti-CD3 or anti-TCR (1-10 g/ml) antibodies. Cells (10 6 /ml) were exposed to an apoptotic stimulus and incubated at 37°C and 5% CO 2 . Apoptotic cells were analyzed by flow cytometry after staining with hypotonic propidium iodide (PI) solution as described previously (25). Briefly, cells (2 ϫ 10 5 ) were incubated for 10 min at 4°C in the dark with 200 l of hypotonic buffer containing 0.1% (w/v) sodium citrate, 0.1% Triton X-100, and 50 g/ml propidium iodide. The cell suspension was then analyzed by flow cytometry, and the subdiploid peak (which represents apoptotic nuclei undergoing DNA fragmentation) was measured to determine the percentage of apoptotic cells.
FasL Functional Activity-FasL functional activity was determined by the ability of FasL expressing cells to induce apoptosis in Fas ϩ LK 35.2 target cells as described previously (6). Briefly, 5 ϫ 10 6 LK 35.2 cells were labeled for 1 h at 37°C with 20 Ci 51 Cr, washed three times, and resuspended in RPMI 1640 medium with 10% fetal calf serum. T cell hybridomas or Jurkat cells were incubated (10 5 cells/well) for 3 h at 37°C in anti-CD3-coated 96-well plates before 51 Cr-labeled LK 35.2 target cells (10 4 cells) were added. Some cells received only 51 Cr-labeled LK35.2 cells and anti-Fas antibodies (Jo2; 5 g/ml). After an additional 8-h incubation, 100 l of supernatant was removed from each well and counted in a ␥-counter to determine experimental release. Spontaneous release was obtained from wells receiving target cells and medium only, whereas total release was obtained from wells receiving 1% Triton X-100. The percentage of specific lysis was calculated with the following formula: percentage of lysis ϭ 100 ϫ ((experimental release Ϫ spontaneous release)/(total release Ϫ spontaneous release)).
Transient Transfection-In transient transfection experiments, J16 (10 7 ) cells were electroporated at 800 microfarads/250 V using a Life Technologies, Inc. electroporator. Cells received 10 g of control plasmid in addition to 10 g of a FasL promoter-luciferase reporter construct, 20 g of experimental construct, and 0.2 g of pRL-TK to normalize for transfection efficiency. Subsequently, cells were incubated for 18 h at 37°C and stimulated with phorbol 12-myristate 13-acetate (PMA; 10 ng/ml) plus ionomycin (2 M) or plate-bound anti-CD3 (1 g/ml) as described above. After 6 h, luciferase activity was measured with a luminometer according to the manufacturer's instructions (Analytical Luminescence, San Diego, CA).

CD3⅐TCR-mediated AICD Involves ERK1/2 Phosphorylation and Activation-
We performed all of our experiments in two T cell lines that are widely used for the study of AICD in T cells: the murine T cell hybridoma DO11.10, which expresses a TCR specific for OVA323-339, a peptide derived from chicken ovalbumin (22), and the human T cell leukemia line Jurkat-16 (a variant of the Jurkat cell line) (4). TCR⅐CD3 cross-linking leads to AICD in both cell lines, whereas Jurkat-16 cells also undergo apoptosis upon Fas cross-linking. In addition, we used in some experiments regular Jurkat cells, which were only sensitive to Fas-mediated apoptosis, or Jurkat-77 cells, which do not undergo AICD upon TCR⅐CD3 stimulation and are resistant to Fas-mediated apoptosis.
We first determined if the ERK pathway is being activated during TCR⅐CD3-mediated AICD as has been described for TCR⅐CD3-mediated activation, proliferation, and IL-2 production (13). Activation of ERK1 and -2 requires phosphorylation of specific tyrosine and threonine residues, and we therefore assessed tyrosine phosphorylation of ERK1 and -2 after CD3 cross-linking in DO11.10 cells by Western blot analysis of whole cell lysates with an antibody that specifically recognizes tyrosine-phosphorylated ERK1 and -2 (tyrosine 204). Maximal phosphorylation of (mostly) ERK2 occurred 15 min after CD3 cross-linking, but ERK1 and -2 remained phosphorylated at later time points (Fig. 1A) up to 6 h (data not shown). Expres-sion of ERK1 and -2 was determined by immunoblotting with anti-ERK antibodies, and expression levels were unaffected by CD3 cross-linking. We also determined with an in vitro kinase assay that the phosphorylation of ERK1 and -2 ( Fig. 1A) correlated with the induction of ERK activity during AICD (Fig.  1B). Subsequently, we analyzed ERK1 and -2 phosphorylation after CD3 cross-linking in two Jurkat subclones: J16 (sensitive to AICD and Fas-mediated apoptosis) and J77 (resistant to AICD and Fas-mediated apoptosis). We found that the kinetics of ERK1 and -2 phosphorylation in these two cell lines was identical (Fig. 1C) and demonstrated a similar pattern as found in DO11.10 cells. All three cell lines demonstrated maximal phosphorylation at 15 min of predominantly ERK2. These data make it unlikely that the kinetics of ERK phosphorylation can account for the differences in susceptibility to AICD seen in these cell lines.
MEK1 Inhibitor PD98059 Inhibits ERK Phosphorylation and Activation during AICD-We then determined if a widely used specific inhibitor of the ERK pathway, MEK1 inhibitor PD98059, could inhibit ERK activation during AICD. PD98059 binds to the inactive form of MEK1 and prevents its activation by upstream activators. This inhibitor has been shown to act as a highly selective inhibitor of the activation of MEK1 in the ERK pathway (26,27). PD98059 inhibits activation and phosphorylation of MEK1 by either c-Raf or MEKK1 with IC 50 values of 5-10 M but inhibits the activation of MEK2 by c-Raf less potently with an IC 50 of 50 M. In all experiments, cells were pretreated with 50 M of PD98059 for 30 min at 37°C. We found that pretreatment with the MEK1 inhibitor PD98059 prevented phosphorylation and activation of ERK1 and -2 completely for at least 2 h after CD3 cross-linking (Fig. 1, A and B). However, phosphorylation and activation of ERK1 and -2 after PMA and ionomycin stimulation was only diminished and not completely inhibited. This partial inhibition could be due to incomplete inhibition of MEK2 by PD98059 or could be the result of an alternative pathway for ERK phosphorylation, which does not involve MEK1 and 2 and/or is not affected by PD98059. Similar results were obtained with Jurkat-16 cells (data not shown).
We also determined the effect of the MEK1 inhibitor PD98059 on the activation of MEK1 upon CD3 stimulation in J16 cells with an antibody that specifically recognizes serine phosphorylation of MEK1 and -2 (serines 217 and 221). Maximal phosphorylation of MEK1/2 occurred after 5 min and could be partially inhibited by the MEK1 inhibitor (Fig. 1D). The incomplete inhibition of MEK1/2 phosphorylation was probably due to the less potent effect of the MEK1 inhibitor against MEK2 activation as described above.
MEK1 Activity Regulates FasL Promotor Activation during AICD-The up-regulation of FasL has been identified as an essential step in TCR⅐CD3-mediated apoptosis in both cell lines (DO11.10 and Jurkat-16) and, in general, during AICD (4 -6). We confirmed the requirement of the Fas/FasL interaction during AICD by inhibiting AICD of DO11.10 T cell hybridomas with Fas-IgFc fusion protein (data not shown). Fas-IgFc binds to FasL and prevents its interaction with the Fas receptor (6). To evaluate the role of the ERK pathway in the regulation of FasL expression during AICD, we co-transfected the human FasL promotor coupled to a luciferase reporter construct along with either a dominant interfering form or a constitutively active form of MEK1, the upstream activator of ERK, in Jurkat-16 cells. CD3 cross-linking resulted in FasL promotor activation, and cotransfection of a dominant-interfering form of MEK1 could inhibit this activation by almost 50% (Fig. 2A). Cotransfection of a constitutively active MEK1 construct led to a 4 -5-fold increase in for 30 min at 37°C and subsequently incubated with plate-bound anti-CD3 antibody. Cells stimulated with PMA (10 ng/ml) and ionomycin (2 M) (P ϩ I) for 15 min at 37°C and 5% CO 2 were analyzed as positive control for ERK phosphorylation. At various intervals, cells were harvested and lysed, and the whole cell lysates were separated by SDSpolyacrylamide gel electrophoresis in 10% gels. ERK phosphorylation (p-ERK1 and p-ERK2) was determined by Western blotting with an anti-phospho-ERK antibody (top panel). Subsequently, the same nitrocellulose membrane was stripped and reprobed with antibodies recognizing ERK1 and -2 to determine ERK1 and -2 expression at every time interval (bottom). A representative experiment out of seven performed is shown. B, DO11.10 T cell hybridoma cells were treated as in A. At various intervals, cells were harvested, and the lysates were immunoprecipitated with an anti-ERK antibody. The immunoprecipitates were tested in an in vitro kinase assay with myelin basic protein as substrate to determine ERK activity. A representative experiment out of three performed is shown. C, J16 and J77 cells were incubated with anti-CD3 antibodies, and at various intervals cells were harvested and lysed, and whole cell lysates were separated as described in A. ERK phosphorylation was determined by Western blotting with anti-phospho-ERK antibody. D, J16 cells were preincubated with 50 M PD98059 MEK1 inhibitor or Me 2 SO (drug vehicle control) for 30 min at 37°C and subsequently incubated with plate-bound anti-CD3 antibody. At various intervals, cells were harvested and lysed, and whole cell lysates were separated as described in A. MEK1/2 phosphorylation was determined by Western blotting with anti-phospho-MEK1/2 antibody. FasL promotor activation in unstimulated cells and a further 6.5-fold increase in FasL promotor activation upon CD3 cross-linking. These results indicate that MEK1 activity can regulate FasL expression during AICD.
The MEK1 Inhibitor PD98059 Inhibits FasL Promotor Activation during AICD-We further studied the role of MEK1 (and the ERK pathway) in FasL expression during AICD with the use of the MEK1 inhibitor PD98059. As shown in Fig. 2B, pretreatment of Jurkat-16 cells with the MEK1 inhibitor could inhibit FasL promotor activation upon CD3 cross-linking. This result confirms our finding with transient transfection of MEK1 constructs that MEK1 activity is required for FasL up-regulation during AICD.
The Inhibition of MEK1 Activity Prevents FasL Expression and Activity after CD3⅐TCR Stimulation-Subsequently, we studied the regulation of FasL expression and activity by MEK1 during AICD. We first determined FasL expression by flow cytometric analysis after intracytoplasmic staining with anti-FasL antibodies. We could not detect any FasL in unstimulated cells (Ͻ1%), whereas a 4-h incubation with anti-CD3 resulted in 13% positive cells. Pretreatment of the cells with Me 2 SO (drug vehicle as control) had no effect on the induction of FasL expression (13% positive cells), whereas pretreatment with the MEK1 inhibitor completely prevented expression of FasL (Ͻ1% positive cells) (Fig. 3A).
We then determined the induction of FasL activity in a cytolytic assay with 51 Cr-labeled Fas-sensitive target cells (6,28,29). This assay measures FasL activity of the effector cells (DO11.10) by their capability to lyse Fas-sensitive target cells (B cell hybridoma LK 35.2). We found that CD3 cross-linking of DO11.10 cells resulted in the induction of FasL activity, which indicates that DO11.10 cells up-regulate their FasL cell surface expression upon CD3⅐TCR activation (Fig. 3B). We confirmed the specificity of this assay by demonstrating that preincubation of CD3-stimulated DO11.10 cells with FasIgFc could inhibit the lysis of the Fas-sensitive target cells (data not shown). Pretreatment of the DO11.10 cells with the MEK1 inhibitor abolished the induction of FasL activity after CD3 cross-linking (Fig. 3B).
AICD Is Inhibited by the MEK1 Inhibitor PD98059 -The use of the MEK1 inhibitor in cell cultures allowed us to test the role of the ERK pathway in CD3-mediated apoptosis assays. Pretreatment of DO11.10 cells with PD98059 resulted in partial inhibition and a 2-4-h delay in the onset of AICD (Fig. 4A). Because PD98059 is only effective for a few hours in cell cultures (see Fig. 1, A and B), we determined if a second addition of the inhibitor would result in a stronger effect. Indeed, the inhibitory effect of PD98059 on AICD of DO11.10 cells could be prolonged and enhanced by a second treatment with PD98059 4 h after the initial treatment (Fig.  4B). We repeated these experiments with Jurkat-16 cells and also found inhibition of AICD after pretreatment with MEK1 inhibitor (Fig. 4C). These results suggest that inhibition of the ERK pathway leads to a marked inhibition of AICD. The inhibitory effect of PD98059 on AICD was confirmed in similar experiments with another murine T cell hybridoma cell line A1.1 (data not shown).
Fas-mediated Apoptosis Is Not Affected by Inhibition of MEK1 Activity-To better evaluate the role of the ERK cascade in the more distal Fas-mediated component of the AICD pathway, we studied Fas-mediated apoptosis of Jurkat cells that resulted from Fas cross-linking (4). We assessed the phosphorylation of ERK1 and -2 during Fas-mediated apoptosis of Jurkat cells by Western blot analysis of whole cell lysates with antibodies against tyrosine-phosphorylated ERK1 and -2. Only very weak, unsustained phosphorylation of ERK2 was found 2 h after Fas cross-linking (Fig. 5A). In addition, we determined activation of ERK1 and 2 after Fas cross-linking of Jurkat cells by in vitro kinase assays. These results were comparable with the analysis of ERK phosphorylation by Western blotting. Very little ERK activation was seen at 2 h after Fas cross-linking, and this activation was not sustained (Fig. 5B). Subsequently, we determined the effect of the MEK1 inhibitor on Fas-mediated apoptosis of Jurkat cells and found no inhibition (Fig. 5C). In addition, we found no inhibition by the MEK1 inhibitor PD98059 of Fas-mediated apoptosis of LK 35.2 cells (Fig. 3B), J16 cells (Fig. 4C), and fresh murine thymocytes (data not shown). These results suggest that activation of the ERK pathway is not required for Fas-mediated apoptosis but is necessary for AICD.
The Anti-apoptotic Effect of the MEK1 Inhibitor PD98059 Is Restricted to AICD-To evaluate the role of the ERK pathway in other forms of apoptosis, we pretreated DO11.10 cells with PD98059 and induced apoptosis by several other stimuli, including corticosteroids (dexamethasone), chemotherapeutic drugs with an inhibitory effect on topoisomerase II (adriamycin and etoposide), C 2 -ceramide, and radiation. We found no inhibitory effect of the MEK1 inhibitor on any of the forms of apoptosis tested (Fig. 6). The anti-apoptotic effect of PD98059 in T cells seems to be specific and restricted to AICD. This suggests that activation of the ERK pathway is required for AICD but not for any of the other forms of apoptosis tested.
The Inhibition of MEK1 Activity Prevents Nur77 Expression during AICD-Liu et al. (19) used the DO11.10 cell line for the identification of an orphan steroid receptor Nur77, which is required for AICD in T cell hybridomas and thymocytes. Nur77 was maximally expressed 2 h after CD3⅐TCR cross-linking of DO11.10 cells. We confirmed CD3⅐TCR-stimulated expression of Nur77 within 2 h by Western blot analysis of whole cell lysates with anti-Nur77 antibodies (Fig. 7). Interestingly, pretreatment of DO11.10 cells with the MEK1 inhibitor prevented this rapid expression of Nur77. In further experiments, we found a 1-2-h delay and overall decrease in Nur77 expression (data not shown). These results suggest that the ERK pathway is required for Nur77 expression during AICD. DISCUSSION In this study, we describe the requirement of signaling through the ERK pathway for AICD of T cells. Several studies have demonstrated activation of the MAPK pathways (ERK, Jun N-terminal kinase, and p38) during apoptosis, and stimulatory or inhibitory effects of these pathways on apoptosis have been found to depend on the cell type, activation or differentiation state, apoptotic stimulus, and cell death pathway. In T cells, the ERK pathway has been linked to various cellular processes, such as IL-2 expression, anergy, activation, and proliferation (13). Our data in AICD of T cells suggest a proapoptotic role for the ERK pathway in AICD in contrast to the antiapoptotic role described for the ERK pathway in apoptosis after nerve growth factor withdrawal (31). Moreover, we also found that pretreatment with MEK1 inhibitor could inhibit AICD of activated nontransformed CD4ϩ splenocytes (data not shown), which confirms the importance of the ERK pathway as a proapoptotic regulator in AICD of T cells. The observation that the ERK pathway can induce apoptosis in certain situations is in agreement with studies by Kauffman-Zeh et al. (32), which demonstrated a proapoptotic role for the ERK pathway in c-Myc-induced apoptosis of fibroblasts after transfection with partial loss-of-function Ras mutants and by Sutherland et al. (33), who found an association between ERK2 activation and B-cell antigen receptor-induced apoptosis in B lymphoma cells. Interestingly, we found that pretreatment with the MEK1 inhibitor would not only inhibit AICD but also inhibited IL-2 secretion (as measured by enzyme-linked immunosorbent assay) after CD3 cross-linking of DO11.10 cells (data not shown). Our data suggest a dual role for the ERK pathway in proliferation and apoptosis in T cells, which has has also been described for c-Myc and CDC2 kinase (1). Green and Scott (1) proposed a two signal death/survival model to explain these results: an activation signal (in this case TCR⅐CD3-mediated ERK activation) leads to proliferation or apoptosis depending FIG. 7. The ERK pathway is required for Nur77 expression during AICD. DO11.10 T cell hybridoma cells were preincubated with 50 M PD98059 MEK1 inhibitor or Me 2 SO for 30 min at 37°C and subsequently incubated with plate-bound anti-CD3 antibody. Some cells were stimulated with PMA (10 ng/ml) and ionomycin (2 M) (P ϩ I) for 15 min at 37°C and 5% CO 2 . At various intervals, cells were harvested and lysed, and the whole cell lysates were separated by SDS-polyacrylamide gel electrophoresis in 10% gels. Nur77 expression was determined by Western blot analysis with anti-Nur77 antibodies. A representative experiment of three experiments is shown. FIG. 5. Fas-mediated apoptosis of Jurkat cells results in weak phosphorylation and activation of ERK1/2 and is not affected by the MEK1 inhibitor. A and B, Jurkat cells were preincubated with 50 M PD98059 MEK1 inhibitor or Me 2 SO and subsequently incubated at 10 6 cells/ml at 37°C with an anti-Fas antibody. Cells stimulated with PMA (10 ng/ml) and ionomycin (2 M) (P ϩ I) for 15 min at 37°C were analyzed as positive control for ERK phosphorylation and activation. At various intervals, cells were harvested, and whole cell lysates were analyzed by Western blotting with an anti-phospho-ERK (p-ERK) antibody (A) to evaluate ERK phosphorylation or by immunoprecipitation with an anti-ERK antibody followed by an in vitro kinase assay with myelin basic protein as substrate (B) to determine ERK activity. C, Jurkat cells were preincubated with 50 M PD98059 MEK1 inhibitor, Me 2 SO, or control medium for 30 min at 37°C and subsequently incubated at 10 6 cells/ml at 37°C with an anti-Fas antibody. Cells were harvested after 10 h, and the percentage of apoptotic cells was determined by flow cytometric analysis after staining with PI solution. A representative experiment out of four experiments performed is shown.
FIG. 6. The MEK1 inhibitor has a specific antiapoptotic effect on AICD and does not affect other forms of apoptosis. DO11.10 T cell hybridomas were preincubated with 50 M PD98059 MEK1 inhibitor or Me 2 SO and subsequently incubated at 10 6 cells/ml for 22 h with plate-bound anti-CD3 antibody, 10 g/ml anti-Fas antibody, 5 M C2ceramide, 10 Ϫ6 M dexamethasone, 10 g/ml adriamycin, 10 g/ml etoposide or exposed to 20-gray ␥-irradiation. The percentage of apoptotic cells was determined by flow cytometric analysis after staining with PI solution.
on additional signals or the cellular context at the time of activation.
Conflicting data exist regarding a possible role of the Ras-ERK pathway in Fas-mediated apoptosis. Goillot (Fig. 3B), and thymocytes (data not shown) was not affected by inhibition of the ERK pathway with MEK1 inhibitor.
NUR77 is an orphan nuclear steroid receptor (17) that belongs to the steroid/thyroid hormone receptor superfamily, which consists of ligand-dependent transcription factors with a characteristic centrally located and highly conserved DNAbinding domain containing two zinc fingers. DNA binding of Nur77 was first detected 2 h poststimulation and continued to increase until 12 h poststimulation. At present, it is not known if NUR77 contributes to the regulation of FasL expression. Expression of FasL in the thymus is highly elevated in NUR77 full-length transgenic mice, which have a substantial decrease in numbers of thymocytes and peripheral T cells, and this would suggest a role for NUR77 expression in the regulation of FasL expression. However, cross-breeding of the NUR77 fulllength transgenic mouse with the gld mouse (which has a mutated FasL gene) only partially rescued NUR77-induced apoptosis (17,39). No ligand for Nur77 has been identified, although in a yeast two-hybrid system using the Nur77 DNA binding domain as a bait a novel cell cycle inhibitor p19 was isolated (40). However, so far no association in vivo between Nur77 and p19 could be demonstrated. Our data suggest that the rapid expression of Nur77 during AICD is regulated by activation of the ERK pathway.
Previous studies have shown that certain signaling molecules, including CD3-, Lck, Zap-70, CD45, Ras, and calcineurin, that are involved in TCR⅐CD3-mediated signaling pathways leading to proliferation and activation are also important for AICD. We have confirmed the requirement of activation of the calcium/calcineurin pathway for AICD, since calcineurin inhibitors (such as cyclosporin A and FK506) can inhibit AICD (data not shown; Ref. 29). Our data with transiently transfected MEK1 mutants and the specific MEK1 inhibitor provide the first evidence that the ERK pathway is required for AICD and exerts its effect through the regulation of FasL expression and perhaps also through the regulation of Nur77 expression. The induction of FasL expression during AICD was shown to be the mechanism by which other signaling molecules (CD3-, Lck, Zap-70, CD45, Ras, and calcineurin) regulate AICD (7-11, 28). However, the effects of these mole-cules on Nur77 expression, which occurs exclusively in T cells and thymocytes during AICD, were not determined in any of these studies. NFAT binding sites have been identified in the FasL promotor (10), which could explain the regulation of FasL expression by the Ca 2ϩ /calcineurin pathway. A binding site for Nur77 in the FasL promotor has not been detected, which leaves the question of how Nur77 regulates AICD unanswered. Also, a binding site for a transcription factor that is regulated by the ERK pathway has not yet been identified within the FasL promotor, although a recently identified MEKK1-regulated response element in the FasL promotor can bind to ATF2 and c-Jun, which are both substrates for phosphorylation by ERK1/2 (41).
In conclusion, our data suggest that the ERK pathway is required for AICD of T cells and the induction of FasL and Nur77 expression during AICD but not for Fas-induced apoptosis of T cells.