Expression of Fas ligand in activated T cells is regulated by c-Myc

Expression of Fas Ligand in Activated T Cells Is Regulated by c-Myc*

Thomas Brunner‡, Shailaja Kasibhatla§^¶, Michael J. Pinkoski§储, Corina Frutschi‡, Nam Jin Yoo§**, Fernando Echeverri§, Artin Mahboubi§, and Douglas R. Green§‡‡

From the§Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, San Diego, California 92121 and‡Division of Immunopathology, Institute for Pathology, University of Berne, 3010 Berne, Switzerland

The transcription factor c-Myc is important for the control of cell cycle progression, neoplasia, and apo- ptotic cell death. c-Myc dimerizes with its partner Max to form an active transcription factor complex. Little is known, however, about the transcriptional targets of c-Myc and their roles in c-Myc-induced cell death. Here we demonstrate that T cell activation-induced expres- sion of Fas ligand (FasL, CD95-L, APO-1-L), which can induce apoptotic cell death in many different cell types, is regulated by c-Myc. Down-modulation of c-Myc pro- tein via antisense oligonucleotides blocked activation- induced FasL mRNA and protein expression and func- tional FasL expression in activated T cells and T cell lines. Further, FasL promoter activity in T cells is driven by overexpression of c-Myc and inhibited by ex- pression of dominant-negative mutants of c-Myc and Max. Our findings indicate that c-Myc controls apo- ptotic cell death in T cells through regulation of FasL expression.

Uncontrolled proliferation of nearly any cell in a complex animal can be catastrophic, and therefore the expansion of cells is tightly regulated. One mechanism of such regulation is the induction of apoptosis, especially by those cellular events that also promote entry into the cell cycle. Thus, c-Myc (1), E2F (2), and adenovirus E1A (3) promote apoptosis in fibroblasts, and this has been suggested to be a critical anticancer mechanism (4). In the immune system, apoptosis of lymphocytes is impor- tant not only as a mechanism to offset oncogenesis but also to prevent damage of host tissues by overly aggressive immune mechanisms (5). Activation of T lymphocytes can therefore promote either cell proliferation or apoptotic cell death (5).

Previous studies using c-Myc antisense oligonucleotides (6) or dominant-negative mutants of c-Myc and Max (7) (which an- tagonize the functional Myc/Max heterodimer) demonstrated that c-Myc function is required for the process of activation-

induced cell death (AICD)¹in T cells. More recently, we (8) and others (9 –11) showed that activation-induced apoptosis in T cells proceeds via expression of Fas (CD95/Apo-1) and its ligand (FasL) and their subsequent interaction on the cell surface.

Similarly, apoptosis triggered by expression of c-Myc in fibro- blasts has also recently been shown to depend on signaling via FasL/Fas interaction on the cell surface (12). Thus, the require- ment for c-Myc in AICD indicates a role for c-Myc in the expression of FasL and/or Fas and/or an increased susceptibil- ity to Fas-mediated apoptosis.

Because AICD appears to be regulated primarily at the level of FasL expression (5), we explored the possibility that FasL expression is controlled by c-Myc. Here we report that inhibi- tion of optimal c-Myc expression by treatment of T cells with antisense oligonucleotides corresponding to c-myc effectively blocked activation-induced expression of FasL mRNA and pro- tein. To examine whether the requirement for c-Myc is at the level of transcription, we employed a reporter construct con- taining 1.2 kb upstream of the human FasL gene (13) and found that this promoter is c-Myc-responsive and -dependent.

These data are discussed in terms of the possibility that the FasL gene may be a direct or indirect transcriptional target for c-Myc.

MATERIALS AND METHODS

Reagents and Cells—The T cell hybridoma lines A1.1, parental L1210, and L1210 expressing murine Fas (L1210.Fas) have been de- scribed previously (8). Human leukemic Jurkat cells were obtained from ATCC. Cells were cultured in RPMI 1640, 5% fetal calf serum, 2 mM L-glutamine, 50 mM␤-mercaptoethanol, 100 units/ml penicillin, 100

␮g/ml streptomycin (complete medium). Normal murine T cell blasts were generated by dissociating the spleens of C57/B6 mice between frosted glass microscope slides and subsequent hypotonic lysis to re- move erythrocytes. After resuspension in complete medium, cells were stimulated for 2 days with 1␮g/ml concanavalin A (Sigma), washed once to remove the lectin, and cultured with 100 units/ml recombinant IL-2 (Proleukin, Chiron) for another 5 days to generate T cell blasts.

Before being used in an experiment, T cell blasts were purified by Ficoll density centrifugation to remove dead cells.

Phorbol 12-myristate 13-acetate (PMA) was purchased from Sigma, and ionomycin was purchased from Calbiochem. Hamster anti-mouse CD3⑀(145-2C11) was purified from culture supernatant by protein A affinity chromatography (14). Normal mouse IgG and biotinylated goat anti-mouse IgG were obtained from Jackson Laboratories (West Grove, PA). Anti-human Fas (CH-11) was obtained from Kamiya (Thousand Oaks, CA). Anti-mouse FasL Kay 10 was a generous gift of Nosheen Alaverdi (Pharmingen, San Diego, CA). The sequence for c-mycanti- sense (AS) oligonucleotides was 5⬘-CACGTTGAGGGGCAT-3⬘, and the sequence for c-mycnonsense (NS) oligonucleotides was 5⬘-AGTGGCG- GAGACTCT-3⬘. All oligonucleotides were phosphorothionate-derivat- ized (Quality Controlled Biochemicals, Hopkinton, MA) (6).

* This research was supported by National Institutes of Health Grant GM52735 (to D. R. G.) and Swiss National Science Foundation Grant 31-54079.98 and Swiss Cancer League Grant SKL 617-2-98 (to T. B.).

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.

¶Present address: Cytovia, 6650 Nancy Ridge Dr., San Diego, CA 92121.

储Supported by a fellowship from the Medical Research Council of Canada.

** Present address: Dept. of Pathology, Catholic University Medical College, 505 Banpo-dong, Socho-gu, Seoul 137-701, South Korea.

‡‡ To whom correspondence should be addressed: Division of Cellular Immunology, La Jolla Institute for Allergy and Immunology, 10355 Science Center Dr., San Diego, CA 92121. E-mail: dgreen5240@aol.com.

1The abbreviations used are: AICD, activation-induced cell death;

FasL, Fas ligand; kb, kilobase(s); IL, interleukin; PMA, phorbol 12- myristate 13-acetate; AS, antisense; NS, nonsense; FACS, fluorescence- activated cell sorter.

9767

Cloning of Human FasL Promoter—We have previously described the cloning of the 1.2-kb FasL promoter (13). Briefly, a 1.2-kb fragment was amplified from pBhFL5H3-1 (15) as template using standard po- lymerase chain reaction conditions. Polymerase chain reactions were done in a Delta II thermocycler from Ericomp (San Diego, CA). Purified DNA was digested withHindIII andSalI and cloned into the luciferase reporter construct HsLuc (16).

Luciferase Assays—Transient transfections of Jurkat cells by elec- troporation and luciferase assay were done as described previously (13).

Briefly, the human FasL HsLuc promoter construct was co-transfected with either (a) pBabe MaxRx, pDOR MycRx, or the combination of both (17) or (b) pSP271 Myc (18). The DNA was normalized with pBabe empty vector to a total amount of 40␮g. For induction of FasL promoter activity, cells were harvested 24 h post-transfection, resuspended in fresh medium, and stimulated with or without PMA (50 ng/ml) and ionomycin (500 ng/ml) for 8 –12 h. Cells were then harvested, washed twice in phosphate-buffered saline, and lysed, and the supernatants were used for detection of luciferase activity with a Monolight 2010 luminometer at the manufacturer’s recommended conditions (Analyti- cal Luminescence Laboratory). All experiments were done in triplicate.

Transfection efficiencies were normalized by co-transfecting with␤-ga- lactosidase expression vector (pSV-␤-Gal, Promega).

Jurkat-Tag cells were suspended at 20⫻10⁶/ml in serum-free media, and 500 cells were transferred to a 4-mm electroporation cuvette. Re- porter construct plasmid DNA was added (15␮g of hFLP-Luc B5 or 10

␮g of NFAT-Luc) along with either empty vector (20␮g of pcDNA3) or c-mycexpression vector (20␮g of pSP271 Myc). After electroporation, which was performed using 960 microfarads and 240 V, cells were cultured for 24 h in complete media and then stimulated with PMA and ionomycin for 16 h. Cells were harvested, washed, and lysed, and luciferase assays were performed as described above and normalized according to protein content.

Assessment of c-Myc Protein—2⫻10⁶A1.1 cells/group were incu- bated with 10 ␮^M c-myc AS or NS for 12 h. After that, cells were harvested, washed once in phosphate-buffered saline, and then lysed in 100␮l of 1⫻SDS-polyacrylamide gel electrophoresis sample buffer (50 mMTris, pH 6.8, 100 mMdithiothreitol, 2% SDS, 10% glycerol). Equal volumes of samples (25␮l) were separated on a 12% gel and transferred to a nylon membrane (polyvinylidene difluoride, Fisher Scientific, Pitts- burgh, PA). Membranes were blocked with 5% nonfat dry milk in Tris-buffered saline, 0.05% sodium azide for 1 h and then incubated with rabbit anti-mouse c-Myc antiserum (1:500) (6) for 4 h at room temperature. Membranes were then washed three times in Tris-buff- ered saline, 0.05% Tween 20 for 15 min and incubated with a donkey anti-rabbit-horse radish peroxidase conjugate (1:2000, Amersham Pharmacia Biotech) for 1 h. After three washes membranes were incu- bated with ECL detection solutions (Amersham Pharmacia Biotech) and exposed to x-ray film. To ensure equal loading, membranes were then stripped and reprobed for actin (mouse anti-actin, 1:2000, Amer- sham Pharmacia Biotech).

Analysis of FasL Expression and Function—Murine T cell hybrido- mas (1⫻10⁷cells/group) were preincubated with 20␮^Mc-mycAS or NS for 4 h and then stimulated with anti-CD3 for 4 h. Total RNA was isolated and analyzed for FasL mRNA by reverse transcriptase-polym- erase chain reaction as described previously (19, 20).

To detect FasL by antibody staining, A1.1 cells were preincubated with 20␮^Mc-mycAS or NS or medium control for 4 h prior to stimu- lation with anti-CD3 for 5 h. After that, cells were harvested; washed in phosphate-buffered saline, 1% calf serum, 0.05% sodium azide (wash buffer); and incubated for 15 min at room temperature with 50␮g/ml normal hamster immunoglobulin in wash buffer to block nonspecific binding. Cells were then placed on ice, and anti-FasL mAb Kay 10 or normal mouse IgG was added (5␮g/ml final concentration). After 30 min cells were washed, and antibody binding was detected with goat anti-mouse biotin (2.5␮g/ml) and streptavidin-fluorescein conjugate (5

␮g/ml), followed by FACS analysis.

To evaluate AICD, cells were pretreated with either 20␮^Mc-mycAS or NS for 4 h prior to stimulation with anti-CD3 for 18 h. Cells were then harvested, and cell death was assessed by propidium iodide uptake (21). No effect of AS or NS c-myc oligonucleotides was observed in anti-Fas-induced apoptosis in Jurkat cells (data not shown).

Activation-induced FasL expression on murine T cells was assessed by determining the ability of these cells to cause DNA fragmentation in Fas^⫹target cells as described previously (8). Briefly, A1.1 cells or T cell blasts were preincubated for 4 h with either medium control, 20␮^M c-mycAS, or 20␮^Mc-mycNS, and then the cells were cocultured with [³H]-thymidine-labeled L1210 or L1210.Fas at an effector:target ratio of 2:1 for A1.1 cells or 4:1 for T cell blasts in the presence or absence of

anti-CD3 stimulation. [³H]-Thymidine-labeled unfragmented DNA was harvested on glass fiber filters and in a liquid scintillation counter.

DNA fragmentation was calculated as follows: % DNA fragmentation⫽ 100⫻(1⫺cpm experimental group/cpm control group)⫾S.D. No DNA fragmentation was observed when target cells were treated with anti- CD3 alone in the absence of T cell hybridomas. Similarly, no DNA fragmentation was observed in the parental cell line L1210 in any of the conditions (data not shown). Assays were performed in triplicate.

RESULTS

Inhibition of c-Myc Expression Blocks Activation-induced FasL but Not Fas Expression in A1.1 T Cells—Our previous studies showed that treatment of the A1.1 T cell hybridoma with antisense oligodeoxynucleotides corresponding to c-myc effectively blocks apoptosis induced by ligation of the T cell receptor on these cells (6), which proceeds via a Fas/FasL interaction (8). We therefore examined the effects of antisense and control oligonucleotides on Fas and FasL expression in these cells. As shown in Fig. 1, treatment with antisense oli- godeoxynucleotides corresponding to c-myc (AS c-myc) inhib- ited c-Myc protein expression (Fig. 1B) and activation-induced apoptosis in these cells (Fig. 1A). Control oligonucleotides with the same base composition but a different sequence (NS c-myc) had no effect in either case. As shown in Fig. 2, treatment of the A1.1 cells with AS c-myc also blocked the induction of FasL mRNA expression (Fig. 2A) and functional FasL expression, as assessed by killing of Fas^⫹ target cells (Fig. 2B). AS c-myc similarly blocked FasL protein expression on activated A1.1 cells as determined by staining with an anti-FasL antibody and FACS analysis (Fig. 2C). Control oligonucleotides (NS c-myc) had no effect on any of the above features. In contrast to FasL expression, treatment of A1.1 cells with AS c-mychad no effect on activation-induced Fas mRNA expression (Fig. 2A). These data support the idea that c-Myc is required for activation- induced FasL but not Fas expression.

To further examine the apparent requirement for c-Myc in activation-induced FasL expression in T cells, we employed normal murine T cells that had been previously activated for 7 days (see “Materials and Methods”). Restimulation of these FIG. 1.Inhibition of Myc protein and AICD in A1.1 cells by AS c-myc.A, inhibition of AICD in A1.1 cells by AS c-myc. A1.1 cells were preincubated with either AS or NS c-mycprior to stimulation with anti-CD3, and cell death was assessed by propidium iodide (PI) uptake after overnight culture.B, inhibition of c-Myc expression by AS c-myc.

A1.1 cells were incubated overnight with either AS or NS c-myc, and c-Myc and actin protein levels in total cell lysates were assessed by Western immunoblot analysis.

cells induced functional FasL expression (Fig. 3A). As we had previously shown in A1.1 cells (6), treatment of the T cells with AS c-myc, but not with NS c-myc, effectively blocked activation- induced apoptosis (Fig. 3A). Further, treatment of activated T cells with AS c-mycprevented activation-induced FasL expres- sion (Fig. 3B). NS c-mychad no effect in either assay.

The FasL Promoter Requires and Is Responsive to c-Myc Expression—To determine whether the requirement for c-Myc for activation-induced expression of FasL is at the level of FasL transcription, we employed a FasL promoter reporter compris- ing 1.2 kb upstream of the human FasL gene (13). When transfected into the Jurkat human leukemic T cell line, the FasL reporter showed a consistent increase in luciferase activ- ity after stimulation with PMA/ionomycin (Fig. 4A). This find- ing is in agreement with the previously described observation that activation of Jurkat cells leads to expression of FasL (9).

If c-Myc transcriptionally regulates FasL, then overexpres- sion of c-Myc should lead to increased FasL reporter activity.

Jurkat cells were therefore co-transfected with the 1.2-kb FasL

promoter reporter construct and a c-Myc expression vector.

Overexpression of c-Myc enhanced both basal and PMA/iono- mycin-induced FasL promoter activity in a gene dosage-de- pendent manner (Fig. 4B). Thus, high levels of c-Myc expres- sion are capable of inducing FasL promoter activity, indicating that c-Myc may play a role in the regulation of FasL gene transcription.

To test the specificity of the effect of c-Myc for the FasL promoter, we compared effects on three promoter reporter con- structs (Fig. 4C). Whereas c-Myc co-expression induced signif- icant induction of the FasL promoter, it had no significant effect on expression of either an IL-2 promoter (not shown) or an NFAT-responsive reporter. Further, this lack of an effect of c-Myc co-expression was also seen upon stimulation of the NFAT reporter by treatment with PMA plus ionomycin. Thus, c-Myc participates in the activation of the FasL promoter but FIG. 2.Inhibition of FasL expression in A1.1 cells by down-

modulation of c-Myc.A, reverse transcriptase-polymerase chain re- action assessment of Fas and FasL expression. A1.1 cells were prein- cubated with either AS or NS c-mycprior to stimulation with anti-CD3.

mRNA expression was detected by reverse transcriptase-polymerase chain reaction as described (19, 20). In the same experiment␤-actin mRNA was measured to ensure equal loading. B, inhibition of func- tional FasL expression by AS c-myc. A1.1 cells were preincubated with either medium control or AS or NS c-mycprior to stimulation with anti-CD3 and co-cultured with radiolabeled target cells. Mean values⫾ S.D. of induction of DNA fragmentation in Fas^⫹target cells are shown.

C, inhibition of cell surface FasL expression by AS c-myc. A1.1 cells were preincubated with either AS or NS c-mycprior to stimulation with anti-CD3. FasL was detected using anti-FasL (Kay10) antibody fol- lowed by FACS analysis.

FIG. 3. Inhibition of FasL function in activated T cells by down-modulation of c-Myc. A, murine T cell blasts were preincu- bated with either AS or NS c-mycprior to stimulation with anti-CD3, and apoptosis was assessed by binding of annexin V-FITC and FACS analysis following overnight culture.B, murine T cell blasts were pre- incubated with either AS or NS c-mycprior to stimulation with anti- CD3 and co-cultured with radiolabeled target cells. Mean values⫾S.D.

of FasL function as measured by induction of DNA fragmentation in Fas^⫹target cells are shown.

FIG. 4.The FasL promoter responds to T cell activation and to c-Myc overexpression.A, activation of the FasL promoter by PMA and ionomycin (P⫹I). Jurkat cells were transfected with the luciferase reporter construct containing a 1.2-kb fragment of the FasL promoter.

Luciferase activity was assessed 12 h post activation with PMA and ionomycin. Arbitrary luciferase units measured of a representative experiment performed in triplicate are shown. B, overexpression of c-Myc enhances human/FasL promoter (hFasLp) activity. Jurkat cells were transiently transfected with 1.2-kb FasL reporter construct and increasing concentrations of c-mycexpression vector as indicated. Lu- ciferase activity (arbitrary units) of triplicate experiments (mean ⫾ S.D.) is shown. C, induction of FasL promoter activity by c-Myc is specific. Jurkat cells were transiently transfected with the indicated reporter plasmid (20␮g)⫾the c-mycexpression vector (40␮g), and the data were standardized as fold induction by c-myccoexpression.

not other irrelevant promoters in this assay.

c-Myc and Max have been found to heterodimerize through their leucine zipper regions at their C-terminal ends, and this complex is responsible for the transcriptional activity (22, 23).

The exchange mutant, MaxRx, in which the leucine zipper region of Max has been exchanged with that of c-Myc, does not bind to c-Myc and can only interact with wild type Max (24) (Fig. 5A). Similarly, its reciprocal mutant MycRx interacts with Myc but does not heterodimerize with Max (24). These hybrids can thereby act as dominant-negative inhibitors of Myc/Max function in different cellular systems (7, 17, 24). We have previously shown that inhibition of Myc/Max function by these dominant-negative mutants prevents activation-induced cell death in T cell hybridomas (7). To assess whether a functional Myc/Max interaction is required for activation-driven induction of the FasL promoter, we co-transfected MaxRx, its reciprocal partner MycRx, or a combination of both into Jurkat cells and analyzed activation-induced promoter activity. Transient transfection of MaxRx consistently led to inhibition of activa- tion-induced luciferase activity (Fig. 5B), indicating a require- ment for Myc/Max in FasL promoter activation. Similarly, transient expression of MycRx significantly inhibited activa-

tion-induced FasL promoter activity, although to a lesser ex- tent than did MaxRx (Fig. 5B). This observation is in agree- ment with previously reported findings that MycRx is a less potent inhibitor of Myc/Max function than is MaxRx (24). Ac- tivation of the FasL promoter followed the predicted pattern (Fig. 5A) as the inhibitory effect of MaxRx or MycRx expression on the FasL promoter activity was abolished upon coexpression of both (Fig. 5B). In contrast, neither of these reciprocal ex- change mutants had any effect on expression of an NFAT- reporter construct (data not shown).

Recently, a Max-interacting protein, Mnt (Fig. 5A), was iden- tified as a repressor of Myc function (25). To determine whether Mnt/Max interactions regulate the FasL promoter, we co-trans- fected Mnt with the 1.2-kb FasL reporter construct. We ob- served that expression of Mnt inhibited activation-induced FasL reporter activity (Fig. 5C).

The FasL Promoter Lacks Consensus c-Myc Binding Se- quences—Myc/Max heterodimers regulate gene transcription via interaction with a defined promoter motif. The canonical sequence CACGTG and its variations as determined by Myc/

DNA binding studies (26) are uniformly absent from the FasL promoter sequence (data not shown). Thus, it is possible that Myc acts indirectly to regulate FasL gene expression via regu- lation of another factor. Alternatively, Myc/Max may bind to a different sequence in the promoter to regulate FasL transcrip- tion. In either case the biological outcome is the same: activa- tion-induced expression of FasL in T cells is dependent on c-Myc.

To partially discriminate between these possibilities, we ex- amined activation-induced FasL mRNA expression in cells treated with cycloheximide to prevent effects of other transcrip- tion factors induced following activation. In the experiment shown in Fig. 6, AS c-myc oligonucleotide pretreatment pre- vented expression of FasL, whereas NS c-mycdid not. Cyclo- heximide did not interfere with this activation-induced, c-Myc- dependent expression of FasL. Thus, transcription and translation of proteins induced by activation (including c-Myc- dependent protein expression) are not required for FasL mRNA expression following activation. This provides support for the idea that Myc/Max acts directly on the FasL promoter to induce FasL expression.

DISCUSSION

When T lymphocytes are optimally activated via ligation of the T cell receptor, they produce cytokines and proliferate, a phenomenon that is central to the induction of an antigen- specific immune response. However, since the original studies of Russellet al.(27) and Lenardo (28), it has been recognized FIG. 5. Activation-induced FasL promoter activity requires

functional Myc/Max heterodimers.A, schematic representation of c-Myc (MycRx) and Max (MaxRx) mutants and the Max-binding Mnt proteins used to inhibit Myc/Max function.LZrepresents the leucine zipper required for heterodimerization; DNA binding domains are in- dicated byboxes.B, effect of co-expression of MycRx and MaxRx on FasL promoter activity. Jurkat cells were transiently transfected with 1.2-kb FasL reporter construct and MaxRx, MycRx, or the combination of both. Mean values of triplicates⫾S.D. of luciferase activity (arbi- trary units) with or without stimulation with PMA/ionomycin (P⫹I) of a representative experiment are shown.C, overexpression of Mnt in- hibits FasL promoter activity. Jurkat cells were transiently transfected with 1.2-kb FasL reporter construct and increasing concentrations of mntexpression vector as indicated. Luciferase activity (arbitrary units) of triplicate experiments (mean⫾S.D.) is shown.

FIG. 6.Activation-induced FasL expression requires c-Myc but not activation-induced proteins. A1.1 cells were activated, and FasL expression was examined by reverse transcriptase-polymerase chain reaction as in Fig. 2. Cells were either pretreated for 6 h with oligonucleotides or treated at the time of activation with cycloheximide (ChX) to inhibit protein synthesis. Cyclosporin A (CsA) was a control for inhibition of FasL expression.

thatrestimulationof a previously activated T cell can result in AICD. Presumably, this phenomenon contributes to immune regulation and the control of activated lymphocyte numbers.

Intuitively, it may be reasoned that because only activated T cells pose a potential threat to immune homeostasis, it is only those cells that are in a state of activation (e.g. expressing c-Myc) that are susceptible to AICD.

How does prior activation contribute to the sensitization of T cells to AICD? Recently, it was shown that previously activated T cells exposed to IL-2 down-regulate c-Flip, an endogenous inhibitor of Fas-mediated apoptosis (29). Thus, an increase in sensitivity to Fas is one aspect of this effect. Here we have shown another: expression of c-Myc in activated T cells contrib- utes to the transcriptional activation of the FasL promoter.

This accounts for the requirement for c-Myc in AICD and provides a mechanism for the relationship between FasL ex- pression and the prior activation status of the T cell.

The observation that c-Myc is required for activation-in- duced FasL expression and AICD in T cells helps to account for the ability of transforming growth factor␤to block this process (30), a phenomenon that may have a role in producing resist- ance of Th2 T cells to AICD and allowing their proliferation (31). Recently, we showed that transforming growth factor␤ inhibits c-Myc expression in T hybridoma cells and that en- forced expression of c-Myc restores AICD and FasL expression (30).

In T cell hybridomas and previously activated T cells, c-Myc expression is constitutive², and such cells are susceptible to activation-induced apoptosis (8 –11), presumably at least in part for this reason. This suggests that activation-induced FasL gene expression requires other transcription factors in addition to Myc/Max. Candidates include calcineurin-activated transcription factors such as NFAT as our previous findings (21) and those of others (9, 32) showed that activation-induced FasL expression is cyclosporin A- and FK506-sensitive. Re- cently, Latiniset al.(33) reported that the region 486 base pairs upstream of the transcription start site in the FasL promoter contains two NFAT sites and that mutating these sites abro- gated FasL reporter activity. In addition, we have recently shown that activation of NF-␬B is important for optimal acti- vation-induced FasL expression, due to an NF-␬B site in the 1.2-kb promoter (34).

Several genes have been shown to be regulated by the “clas- sical” Myc E-box containing enhancer elements including pro- thymosin␣, ornithine decarboxylase, elongation initiation fac- tors 2a and 4E, p53, cyclin D1, and Cdc25 (35– 42). The ornithine decarboxylase gene is directly induced by the Myc/

Max heterodimer through two c-Myc recognition sites in its first intron. Overexpression of ornithine decarboxylase in cell lines is transforming (43) and also appears to mediate c-Myc- mediated susceptibility to apoptosis induced by IL-3 with- drawal (44). Another recently identified target gene of the c-Myc/Max heterodimer iscdc25A, a cyclin-dependent, kinase- activating phosphatase that plays a crucial role in cell cycle progression. c-Myc/Max heterodimers bind directly to the cdc25A promoter through Myc/Max consensus sequences and activate transcription (45). Cdc25A can also mediate transfor- mation and appears to promote c-Myc-induced apoptosis in fibroblasts. In general, these studies point to a role for cell cycle progression in some forms of c-Myc-induced apoptosis, and some investigators have therefore suggested that such apopto- sis is caused by inappropriate entry into the cell cycle through a cell cycle checkpoint (46). Our finding that c-Myc regulates the expression of FasL in T cells connects c-Myc to apoptotic

cell death through engagement of the Fas-mediated apoptotic pathway. Whether any of these genes or another gene is re- sponsible for indirectly activating the FasL promoter is not known. Alternatively, c-Myc may act directly on the promoter via a noncanonical site.

In addition to activation-induced cell death in T cells, fibro- blasts are probably the most extensively studied system for Myc-induced apoptosis (1). The recent finding that Myc-in- duced apoptosis in serum-starved fibroblasts is dependent on Fas/FasL interactions (12) helps to support our finding that fasLis a target gene of c-Myc. These authors, however, were reportedly unable to measure the very low levels of FasL ex- pression in these cells and did not observe changes upon c-Myc expression.

Cell types other than T cells are capable of expressing FasL.

Of particular interest is the finding that tumor cells such as colon carcinomas (47), melanomas (48), and lung carcinomas (49) can express FasL in a constitutive manner. Although we expect that these transformed cells constitutively express c- Myc, they presumably lack other T cell receptor-derived sig- nals. It is possible that c-Myc expression in these cell types is sufficient for FasL expression or that activation-induced sig- nals are replaced by others. Detailed analysis of FasL expres- sion in these cell types should bring further insight into the alternative regulation of FasL expression in non-T cells. Many tumor cells are insensitive to Fas cross-linking and/or express low levels of Fas (47, 48). Nevertheless, FasL expression on tumor cells can have immunological consequences resulting either in inhibition of antitumor immunity (47– 49) or FasL- induced granulocytosis (50). Thus, the inhibition of c-Myc ex- pression in such cells might effectively down-regulate FasL expression, providing an attractive route to the generation of effective antitumor responses.

Acknowledgments—We thank B. Amati for c-MycRx and MaxRx con- structs; S. Nagata for mouse FasL cDNA and a human genomic FasL clone; R. N. Eisenman formycandmaxexpression vectors, antibodies, and glutathioneS-transferase constructs; P. Golstein for mouse Fas cDNA and the L1210.Fas cell line; W. Force for HsLuc vector; N.

Alaverdi for Kay10 anti-FasL; and R. Kluck and C. Mueller for critical reading of the manuscript.

REFERENCES

1. Evan, G. I., Wyllie, A. H., Gilbert, C. S., Littlewood, T. D., Land, H., Brooks, M., Waters, C. M., Penn, L. Z., and Hancock, D. C. (1992)Cell69,119 –128 2. Wu, X., and Levine, A. J. (1994)Proc. Natl. Acad. Sci. U. S. A.91,3602–3606 3. White, E., Cipriani, R., Sabbatini, P., and Denton, A. (1991)J. Virol.65,

2968 –2978

4. Harrington, E. A., Fanidi, A., and Evan, G. I. (1994)Curr. Opin. Genet. Dev.4, 120 –129

5. Brunner, T., Yoo, N. J., Griffith, T. S., Ferguson, T. A., and Green, D. R. (1997) Behring Inst. Res. Commun.97,161–174

6. Shi, Y., Glynn, J. M., Guilbert, L. J., Cotter, T. G., Bissonnette, R. P., and Green, D. R. (1992)Science257,212–214

7. Bissonnette, R. P., MacGahon, A. J., and Green, D. R. (1994)J. Exp. Med.180, 2413–2418

8. Brunner, T., Mogil, R. J., LaFace, D., Yoo, N. J., Machboubi, A., Echeverri, F., Martin, S. J., Force, W. R., Lynch, D. H., Ware, C. F., and Green, D. R.

(1995)Nature373,441– 444

9. Dhein, J., Walczak, H., Baeumler, C., Debatin, K.-M., and Krammer, P. H.

(1995)Nature373,438 – 441

10. Alderson, M. R., Tough, T. W., Davis-Smith, T., Braddy, S., Falk, B., Schooley, K. A., Goodwin, R. G., Smith, C. A., Ramsdell, F., and Lynch, D. H. (1995) J. Exp. Med.181,71–77

11. Ju, S.-T., Panka, D. J., Cui, H., Ettinger, R., El-Khatib, M., Sherr, D. H., Stanger, B. Z., and Marshak-Rothstein, A. (1995)Nature373,444 – 448 12. Hueber, A. O., Zornig, M., Lyon, D., Suda, T., Nagata, S., and Evan, G. I. (1997)

Science278,1305–1309

13. Kasibhatla, S., Brunner, S., Genestier, G., Echeverri, E., Mahboubi, A., and Green, D. R. (1998)Mol. Cell1,543–551

14. Leo, O., Foo, M., Sachs, D. H., Samelson, L. E., and Bluestone, J. A. (1987) Proc. Natl. Acad. Sci. U. S. A.84,1374 –1378

15. Takahashi, T., Tanaka, M., Inazawa, J., Abe, T., Suda, T., and Nagata, S.

(1994)Int. Immunol.6,1567–1574

16. Tillman, J. B., Crone, D. E., Kim, H. S., Sprung, C. N., and Spindler, S. R.

(1993)Mol. Cell. Endocrinol.95,101–109

17. Amati, B., Littlewood, T. D., Evan, G. I., and Land, H. (1993)EMBO J.12, 5083–5087

18. Kretzner, L., Blackwood, E. M., and Eisenman, R. N. (1992)Nature359,

2T. Brunner and D. R. Green, unpublished results.

426 – 429

19. Bonfoco, E., Stuart, P. M., Brunner, T., Lin, T., Griffith, T. S., Gao, Y., Nakajima, H., Henkart, P. A., Ferguson, T. A., and Green, D. R. (1998) Immunity9,711–720

20. Genestier, L., Brunner, T., Kasibhatla, S., and Green, D. R. (1999)J. Exp. Med.

189,231–239

21. Brunner, T., Yoo, N. J., LaFace, D., Ware, C. F., and Green, D. R. (1996)Int.

Immunol.9,1017–1026

22. Blackwell, T. K., Kretzner, L., Blackwood, E. M., Eisenman, R. N., and Weintraub, H. (1990)Science250,1149 –1151

23. Blackwood, E. M., and Eisenman, R. N. (1991)Science251,1211–1217 24. Amati, B., Brooks, M. W., Levy, N., Littlewood, T. D., Evan, G. I., and Land, H.

(1993)Cell72,233–245

25. Hurlin, P. J., Queva, C., and Eisenman, R. N. (1997)Genes Dev.11,44 –58 26. Grandori, C., Mac, J., Siebelt, F., Ayer, D. E., and Eisenman, R. N. (1996)

EMBO J.15,4344 – 4357

27. Russell, J. H., White, C. L., Loh, D. Y., and Meleedy-Rey, P. (1991)Proc. Natl.

Acad. Sci. U. S. A.88,2151–2155 28. Lenardo, M. J. (1991)Nature353,858 – 861

29. Refaeli, Y., Van Parijs, L., London, C. A., Tschopp, J., and Abbas, A. K. (1998) Immunity8,615– 623

30. Genestier, L., Kasibhatla, S., Brunner, T., and Green, D. R. (1999)J. Exp. Med.

189,231–239

31. Zhang, X., Giangreco, L., Broome, H. E., Dargan, C. M., and Swain, S. L. (1995) J. Exp. Med.182,699 –709

32. Anel, A., Buferne, M., Boyer, C., Schnitt-Verhulst, A.-M., and Golstein, P.

(1994)Eur. J. Immunol.24,2469 –2476

33. Latinis, K. M., Carr, L. L., Peterson, E. J., Norian, L. A., Eliason, S. L., and Koretzky, G. A. (1997)J. Immunol.158,4602– 4611

34. Kasibhatla, S., Genestier, L., and Green, D. R. (1999)J. Biol. Chem.274, 987–992

35. Eilers, M., Schirm, S., and Bishop, J. M. (1991)EMBO J.10,133–141 36. Bello-Fernandez, C., Packham, G., and Cleveland, J. L. (1993)Proc. Natl.

Acad. Sci. U. S. A.90,7804 –7808

37. Rosenwald, I. B., Rhoads, D. B., Callanan, L. D., Isselbacher, K. J., and Schmidt, E. V. (1993)Proc. Natl. Acad. Sci. U. S. A.90,6175– 6178 38. Roy, B., Beamon, J., Balint, E., and Reisman, D. (1994)Mol. Cell. Biol.14,

7805–7815

39. Daksis, J. I., Lu, R. Y., Facchini, L. M., Marhin, L. M., and Penn, L. J. Z. (1994) Oncogene9,3635–3645

40. Hermeking, H., and Eick, D. (1994)Science265,2091–2093

41. Gaubatz, S., Imhof, A., Dosch, R., Werner, O., Mitchell, P., Buettner, R., and Eilers, M. (1995)EMBO J.14,1508 –1519

42. Jones, R. M., Branda, J., Johnston, K. A., Polymenis, M., Gadd, M., Rustgi, A., Callanan, L., and Schmidt, E. V. (1996)Mol. Cell. Biol.16,4754 – 4764 43. Auvinen, M., Paasinen, A., Andersson, L. C., and Holtta, E. (1992)Nature360,

355–358

44. Packham, G., and Cleveland, J. L. (1994)Mol. Cell. Biol.14,5741–5747 45. Galaktionov, K., Chen, X., and Beach, D. (1996)Nature382,511–517 46. Bissonnette, R. P., Shi, Y., Mahboubi, A., Glynn, J. M., and Green, D. R. (1994)

inApoptosis II: The Molecular Basis of Apoptosis in Disease(Tomei, L. D., and Cope, F. O., eds) pp. 327–356, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

47. O’Connell, J., O’Sullivan, G. C., Collins, J. K., and Shanahan, F. (1996)J. Exp.

Med.184,1075–1082

48. Hahne, M., Rimoldi, D., Schroter, M., Romero, P., Schreier, M., French, L. E., Schneider, P., Bornard, T., Fontana, A., Lienhard, D., Cerottini, J., and Tschopp, J. (1996)Science274,1363–1366

49. Niehans, G. A., Brunner, T., Frizelle, S. P., Liston, J. C., Salerno, C. T., Knapp, D. J., Green, D. R., and Kratzke, R. A. (1997)Cancer Res.57,1007–1012 50. Yagita, H., Seino, K., Kayagaki, N., and Okumura, K. (1996)Nature379,682