The Human Tumor Necrosis Factor (TNF) Receptor-associated Factor 1 Gene (TRAF1) Is Up-regulated by Cytokines of the TNF Ligand Family and Modulates TNF-induced Activation of NF-κB and c-Jun N-terminal Kinase*

To understand how the TNF receptor-associated factor 1 (TRAF1) is transcriptionally regulated, in vitroDNA binding assays, promoter-reporter gene assays, and RNase protection assays were performed with the human TRAF1 gene. Binding of NF-κB to three of five putative binding sites within the human TRAF1 promoter was found in electrophoretic mobility shift assay studies, and analysis of TRAF1 gene promoter luciferase constructs confirmed the functional importance of these elements. Moreover, triggering of TNF-R1, CD40, and the interleukin-1 receptor resulted in transcription of the TRAF1 gene, whereas receptors that are not activators or only poor activators of NF-κB in HeLa cells failed to show a significant TRAF1 induction. Because it has been shown that members of the TRAF family are involved in activation of NF-κB and the c-Jun N-terminal kinase (JNK) by the interleukin-1 receptor and members of the TNF receptor superfamily, a role of TRAF1 in receptor cross-talk and/or feedback regulation of activated receptor signaling complexes can be suggested. In fact, we found that TNF-induced activation of JNK is prolonged in transfectants overexpressing TRAF1, whereas overexpression of a deletion mutant of TRAF1 in which the N-terminal part had been replaced by the green fluorescent protein interfered with TNF-induced activation of NF-κB and JNK.

proteins serve as common adapter proteins that connect members of the TNF receptor superfamily and the IL-1-R to pathways leading to the activation of NF-B and the c-Jun Nterminal kinase (JNK). To date six different members of the TRAF family have been identified in man and mouse. All TRAF molecules contain the conserved C-terminal TRAF domain, which mediates homo-and hetero-oligomerization, receptor binding, as well as association with a number of cytoplasmic proteins including cIAP1 (cellular inhibitor of apoptosis protein 1; Ref. 21), cIAP2 (21), A20 (22), TRIP (TRAF-interacting protein; Ref. 23 (29), and NIK (NF-B inducing kinase; Refs. 9 and 30). For most of these molecules a role in regulation of NF-B as well as JNK activation and apoptosis has been suggested. Interestingly, TRAF2 knockout mice, as well as mice overexpressing a dominant negative mutant of TRAF2, are functional in TNF-dependent NF-B activation but not in activation of JNK (31,32). To date, it is unclear whether the functionality of the TNF-initiated pathway leading to NF-B activation in these mice is caused by compensatory effects of other TRAFs or reflects a dispensability of TRAF2 for TNF-induced NF-B activation (31,32). Conversely, RIP knockout mice are fully deficient in NF-B activation but exert a normal JNK response upon TNF treatment (33).
The N-terminal region of most of the TRAF proteins contains a RING finger motif and an additional array of zinc finger structures that are arranged in two or three CART domains (34). TRAF2, 3, and 6 are constitutively expressed in various tissues (3,11,12,15,18), whereas the expression of TRAF1, 5, and 4 is more restricted (11,15,17,34). TRAF1 is unique among the TRAF molecules in two aspects; first, it contains no RING finger motif, and second, the zinc finger structures of TRAF1 are not arranged in CART domains. TRAF1 mRNA levels are about tenfold increased in Epstein-Barr virus-positive cells compared with noninfected cells (15) and can be induced in T lymphocytes by stimulation of the T-cell receptor complex (19). TRAF1 was originally cloned as a factor that is recruited to the TNF-R2 signaling complex via the TRAF2 molecule (11) but associates also directly with CD30 (19,20), herpesvirus entry mediator (7), 4 -1BB (8), and the Epstein-Barr virus protein LMP1 (latent membrane protein-1; Ref. 14). In some recent reports, the role of TRAF1 in LMP1-and CD30mediated NF-B activation (4,35), as well as in antigen-induced apoptosis of CD8 ϩ T lymphocytes, has been discussed (36). Here we show that the TRAF1 promoter contains several functional NF-B sites. In accordance with these data, we can demonstrate that stimulation of the NF-B inducing receptors  TNF-R1, TNF-R2, CD40, TCR, and IL-1-R activates the transcription of TRAF1. Because overexpression of a deletion mutant of TRAF1 significantly reduces TNF-R1-induced activation of NF-B and JNK, we propose that TRAF1 has a positive modulatory role in cytokine-induced gene induction.

EXPERIMENTAL PROCEDURES
Materials-293 human embryonic kidney cells were a generous gift from Dr. Gary J. Nabel (Howard Hughes Medical Institute, University of Michigan, Ann Arbor, MI). Recombinant human TNF was purchased from Sigma or was a gift from I.-M. von Broen (Knoll AG, Ludwigshafen, Germany). The TNF-R2-specific agonistic monoclonal antibody MR2-1 was kindly provided by W. Buurman (University of Limburg, Maasstricht, the Nederlands). All other reagents were obtained from Sigma if not otherwise stated. A 1:1 mixture of purified recombinant Flag-tagged human TRAIL (10 g/ml) and anti-Flag M2 antibody (20 g/ml; Kodak International Biotechnologies) was incubated at room temperature for 10 min and diluted to the final concentrations indicated in the figure legends. A cDNA fragment comprising human TRAF1(185-408) with a 5Ј BamHI site and a 3Ј NotI site was generated using proofreading polymerase chain reaction with the p-RSK-TRAF1 plasmid as template. The BamHI and NotI sites were used for subsequent cloning in the recently described pcDNA3.1-GFP⌬FADD construct (37) whereby the ⌬FADD part was previously removed. The TRAF1 promoter-reporter gene construct pGL3-T1p(Ϫ1 to Ϫ1404) was cloned by ligation of a 1.4-kilobase pair fragment of genomic TRAF1 into the blunted XhoI site of the pGL3 basic plasmid (Promega) immediately 5Ј of the luciferase gene (numbering corresponds to bases of the TRAF1 promoter in GenBank TM accession number Y10284). pGL3-T1p(Ϫ1 to Ϫ424) was constructed by deletion of an NdeI/EcoRI fragment, blunt ending, and religation of pGL3-T1p(Ϫ1 to Ϫ1404). To generate pGL3-T1p(Ϫ1 to Ϫ31), the plasmid pGL3-T1p(Ϫ1 to Ϫ1404) was digested with SacI and religated. The pGL3-T1p(Ϫ1 to Ϫ202) and pGL3-T1p(Ϫ1 to Ϫ121) plasmids were cloned by polymerase chain reaction using sense 5Ј-GGG GTA CCA CTC CTA AAG CCT TCA GTC-3Ј or sense 5Ј-GGG GTA CCA ACA AAG GGT AAT TCCTGC-3Ј and antisense 5Ј-ATG CCA AGC TTA CTT AGA TC-3Ј oligonucleotides and the TRAF1 promoter as template. Products were digested with KpnI and HindII and ligated into pGL3. Amplified products and cloning junctions were verified by DNA sequencing.
Site-directed Mutagenesis of pGL3-T1p(Ϫ1 to Ϫ1404)-Mutations were introduced into pGL3-T1p(Ϫ1 to Ϫ1404) using the Quick Change Site-directed Mutagenesis Kit (Stratagene) according to the manufacturer's recommendations. In brief, two complementary primers comprising the mutated B sites B1, B3, and B5 of the human TRAF1 promoter (B1-mut sense, 5Ј-GGA CTT GTC TCC AAC ACC CCT CGA GTT TCC ACC AGG AAG GTG AGC-3Ј; B1-mut antisense, 5Ј-GCT CAC CTT CCT GGT GGA AAC TCG AGG GGT GTT GGA GAC AAG TCC-3Ј; B3-mut sense, 5Ј-GCC TGC GAT TCT CAA CCA GCT CGA GTC TCA CTG TGC TTT CTG AGA G-3Ј; B3-mut antisense, 5Ј-CTC TCA GAA AGC ACA GTG AGA CTC GAG CTG GTT GAG AAT CGC AGG C-3Ј; B5-mut sense, 5Ј-CAG GGG ATT TTT ATC GCA ACA AAG CTC GAG TCC TGC TCC ATC CCT GCT G-3Ј; B5-mut antisense, 5Ј-CAG CAG GGA TGG AGC AGG ACT CGA GCT TTG TTG CGA TAA AAA TCC CCT G-3Ј; mutated B sites with introduced XhoI site are underlined) were extended during temperature cycling by means of Pfu DNA polymerase resulting in mutated plasmids containing staggered nicks. The products are then treated with DpnI, which selectively digests methylated (parental) and hemimethylated (semi-parental) plasmids. The nicked, mutation-containing plasmids were then transformed into Escherichia coli XL1-Blue, and several clones were analyzed for the presence of the desired mutation. The introduced mutation creates an additional XhoI restriction site allowing the identification of mutant plasmids by restriction analysis. Finally, the mutants were controlled by sequence analysis.
Cell Culture and DNA Transfections-HeLa cells were maintained in Clicks/RPMI 1640 medium supplemented with 5% heat-inactivated fetal calf serum, 2 mM L-glutamine, 50 units/ml penicillin, and 50 g/ml streptomycin. 293 cells were grown in Dulbecco's modified Eagle's medium that was supplemented with 10% heat-inactivated fetal calf serum, 2 mM L-glutamine, 50 units/ml penicillin, and 50 g/ml streptomycin. One day prior to transfection, cells were plated at a density of 1.5 ϫ 10 6 /100-mm tissue culture dish. Stable TRAF1 transfectants were obtained by cotransfection of the TRAF1 expression construct p-RSK-TRAF1 encoding full-length human TRAF1 and pcDNA3.1 containing the neomycin resistance gene for G418 selection with SuperFect according to the manufacturer's recommendations (Qiagen, Hilden, Germany). Stable clones were obtained after 2 weeks of selection, and individual drug-resistant colonies were isolated, expanded, and characterized. HeLa clones overexpressing GFP-TRAF1(185-408) were also generated by Superfect based transfection. Transfected cells were selected in culture medium containing 600 g/ml G418 (Life Technologies, Inc.). 20 days post-transfection individual drug-resistant colonies were pooled, expanded, and enriched for high expression by three cycles of cell sorting using a FACStarplus (Becton Dickinson, San Jose, CA).
Electrophoretic Mobility Shift Assay-Nuclear protein extracts were prepared essentially as described by Dignam et al. (38) with some modifications (39). In brief, cells were washed two times with phosphate-buffered saline and resuspended in 0.32 M sucrose, 2 mM magnesium acetate, 3 mM CaCl 2 , 0.1 mM EDTA, 10 mM Tris-HCl, pH 8.0, 0.5% Nonidet P-40, 1 mM DTT, and 0.5 mM PMSF, and nuclei were separated by centrifugation (2600 rpm, 5 min, 4°C). Glycerol was added to a final concentration of 25%, and extracts were stored at Ϫ70°C. Nuclei were washed once in 0.32 M sucrose, 2 mM magnesium acetate, 3 mM CaCl 2 , 0.1 mM EDTA, and 10 mM Tris-HCl, pH 8.0, and resuspended in low salt buffer (0.02 KCl, 0.2 mM EDTA, 1.5 mM MgCl 2 , 20 mM Hepes, pH 7.9, 0.5 mM DTT, 0.5 mM PMSF, and 25% glycerol; 20 l for 1 ϫ 10 7 cells or 40 l for 5 ϫ 10 7 cells). After 5 min incubation the same volume of high salt buffer (0.8 KCl, 0.2 mM EDTA, 1.5 mM MgCl 2 , 20 mM Hepes, pH 7.9, 0.5 mM DTT, 0.5 mM PMSF, and 25% glycerol) was added in four fractions. After 20 min of incubation at 4°C, 1.5 volumes of 0.1 mM EDTA, 0.5 DTT, 0.5 mM PMSF, 25 mM Hepes, pH 7.9, and 25% glycerol were added. Nuclei were centrifuged (14000 rpm, 15 min, 4°C), and aliquots of nuclear protein extracts were stored at Ϫ70°C. Protein concentrations were determined by the method of Bradford (Ref. 50; Bio-Rad). The oligonucleotide probe used for electrophoretic mobility shift assay corresponded to the high affinity B sequences found in the mouse light chain enhancer and in the human immunodeficiency virus promoter region. Two oligonucleotides were annealed to generate a double-stranded probe: sense 5Ј-AGC TTG GGG ACT TTC CAC TAG TAC G-3Ј and antisense 5Ј-AAT TCG TAC TAG TGG AAA GTC CCC A-3Ј (the binding sites are underlined). End labeling was accomplished by treatment with T4 kinase in the presence of [␥-32 P]ATP. Labeled oligonucleotides were purified on push columns (Stratagene). Labeled double-stranded probe (80,000 cpm) was added to 6 g of nuclear protein in the presence of 1 g of poly(dI⅐dC) as nonspecific competitor (Amersham Pharmacia Biotech). Binding reactions were carried out in 10 mM Tris-HCl, pH 7.5, 100 mM NaCl, 4% glycerol for 30 min at 4°C. DNA protein complexes were resolved by electrophoresis on a 4% nondenaturating polyacrylamide gel in Tris/glycine/EDTA buffer. Gels were vacuum dried and exposed to Kodak XAR-5 film at Ϫ70°C with intensifying screens. To identify the binding properties of potential NF-B/Rel-binding sites in the TRAF1 promoter region, competition experiments were performed by adding specific unlabeled doublestranded oligonucleotides to the reaction mixture in 20-or 100-fold molar excess. Oligonucleotides spanning putative B sites of the TRAF1 promoter region were annealed to generate double-stranded competitors and shown in Fig. 1A.
Transient Transfections of Plasmids to HeLa Cells and Luciferase Assay-Transfections for analysis of the TRAF1 promoter reporter constructs were performed using the calcium phosphate precipitation method as described previously (39). Briefly, HeLa cells were seeded 24 h before transfection (5 ϫ 10 5 cells/well) in 6-well dishes. For transfection 10 g of plasmid-DNA was mixed with 124 M CaCl 2 in a total volume of 300 l. After 18 h, the culture medium was removed, and cells were washed once with phosphate-buffered saline and stimulated for 24 h with TNF (150 units). Cell extracts were prepared by the addition of 200 l of cell lysis buffer and one freeze-thaw cycle (Luciferase Assay System, Promega). A portion of the extracts (10 l) was added to luciferase substrate (50 l). The luminescence was quantitated in the single photon mode using a Berthold luminometer. Background from the substrate solution and the counter was 90 -100 relative light units. 293 cells transfected with the pGL3 basic construct alone usually revealed 200 -500 light units, which were taken as basal activity. Transfections were performed in at least four independent experiments. Data are presented as the means Ϯ S.E. For NF-B reporter plasmid assays HeLa cells (0.8 ϫ 10 5 ) were seeded in 24-well tissue culture plates. The following day the cells were transfected using the SuperFect reagent (Qiagen) with a 3ϫ NF-B-luciferase reporter plasmid and a SV40 promoter-driven ␤-galactosidase expression plasmid to normalize the transfection efficiency. After additional 24 h, the cells were harvested in phosphate-buffered saline, lysed in luciferase lysis buffer (Promega, Mannheim, Germany), and assayed for luciferase and ␤-galactosidase activities using an LUMAT 9501 Luminometer (Berthold, Bad Wildbad, Germany).
RNase Protection Assay-Cells (8 ϫ 10 6 ) were plated on a 150-mm Petri dish in 12 ml of Clicks/RPMI 1640 medium overnight at 37°C. The next day the cells were treated with the reagent of interest, and subsequently RNA was isolated with RNA INSTAPURE (Eurogentech, Seraing, Belgium) according to the manufacturer's recommendations. Then the RNAs were tested for the presence of transcripts of xIAP, TRAF1, TRAF2, TRAF4, NAIP, cIAP2, cIAP1, TRPM2, and TRAF3 using the hApo-5 Multi-Probe template set (PharMingen, Hamburg, Germany). In this set L32 and GAPDH were included as internal controls. Probe synthesis, hybridization and RNase treatment were performed with the RiboQuant Multi-Probe RNase Protection Assay System (PharMingen) according to the manufacturer's recommendations. Finally, samples were analyzed by electrophoresis on denaturing polyacrylamide gels (5%) and quantified by phosphorimaging.
Immunocomplex JNK Assay-Jun kinase assays were performed upon immunoprecipitation of JNK1 using a rabbit antiserum (Santa Cruz Biotechnology). GST-Jun was used as substrate in an in vitro kinase assay as described elsewhere (40).

RESULTS AND DISCUSSION
The Putative Promoter Region of TRAF1 Is Inducible by TNF and Contains Three Functional B Sites-In a recent publication, we have described the molecular cloning of the promoter of the human TRAF1 gene (41). To identify cis-acting sequences that regulate transcription of TRAF1, a series of reporter plasmids was constructed. Different 5Ј deleted fragments of the putative TRAF1 promoter region were ligated immediately upstream of the luciferase reporter gene in the pGL3 basic plasmid (Fig. 1B). Transient transfections into HeLa cells were performed. Transfections of pGL3-T1p(Ϫ1 to Ϫ1404) and pGL3-T1p(Ϫ1 to Ϫ424) showed a significant basal activity which dropped by further deletion of 222 (pGL3-T1p(Ϫ1 to Ϫ202)) and 303 base pairs (pGL3-T1p(Ϫ1 to Ϫ121)), respectively. An additional deletion of 90 base pairs yielded the pGL3-T1p(Ϫ1 to Ϫ31) construct, which had almost completely lost its basal activity (Fig. 1C). Transcriptional activity of the pGL3-T1p(Ϫ1 to Ϫ1404) construct could be increased 4.8-fold by stimulation with TNF. The respective levels of induction by TNF decreased stepwise when transfecting the different 5Ј deleted TRAF1 promoter constructs (Fig. 1D). Sequence analysis revealed five putative NF-B/Rel consensus sites in the TRAF1 promoter region (Fig. 1, A and B). To get first insights in the functional importance of these B elements, we tested whether NF-B/Rel proteins bind to these sites. Hence, electrophoretic mobility shift assay analyses were performed using unlabeled oligonucleotides encompassing the different putative NF-B/Rel sites as competitors. The classical B site of the murine light chain enhancer (IgB) was used as probe and incubated with nuclear extracts from TNF stimulated 293 cells. NF-B/Rel complexes bound to the IgB site efficiently and were competed by addition of a 20-or 100-fold molar excess of unlabeled IgB oligonucleotide (Fig. 1E). Binding specificity was controlled with the unrelated Sp1 site, which did not interfere with NF-B/Rel binding. Similar to unlabeled IgB, oligonucleotides containing the 1, 3, and 5 B sites of the TRAF1 promoter region competed for NF-B/Rel binding. In contrast, the 2 and 4 sites did not bind NF-B/Rel with high affinity and are therefore not likely to be relevant for NF-B/ Rel-mediated transcription. To further substantiate the functional importance of the B sites B1, B3, and B5, we individually disrupted these sites in the TRAF1 promoter by sitedirected mutagenesis. As shown in Fig. 1F, TNF-mediated induction of a TRAF1 promoter-driven reporter gene is significantly impaired when the B site B1 or B5 is disrupted, whereas mutation of B3 has only a minor effect on TNF-dependent up-regulation of the reporter gene. A dominant role of the B sites B1 and B5 in TNF-dependent up-regulation of a TRAF1 promoter-driven reporter gene is consistent with the sequential decrease in TNF-mediated TRAF1 promoter activity found upon removal of these sites by deletion mutagenesis (Fig.  1D). In addition, deletion mutagenesis (Fig. 1D), as well as site-directed mutagenesis (Fig. 1F), argues for a minor role of the B site B3, although an oligonucleotide comprising this site efficiently competes for binding to an IgB oligonucleotide (Fig. 1E).

NF-B-inducing Factors Up-regulate TRAF1
Transcription-It has been shown that both TNF receptors are capable of mediating activation of NF-B. Therefore, we investigated the effect of TNF on TRAF1 transcription with the RNase protection technique using a commercially available multitemplate set containing specific probes for TRAF1, TRAF2, TRAF3, TRAF4, NAIP, cIAP1, cIAP2, xIAP, and TRPM2 as well as L32 and GAPDH as internal controls. As shown in Fig. 2A we found a dose-dependent induction of TRAF1 message in various TNFresponsive cell lines. Interestingly, cIAP2, which was recently identified as a NF-B-dependent gene (42), was also up-regulated upon TNF treatment. In contrast, TNF had no effect on the transcription of the other members of the TRAF and IAP family detectable with this template set. HeLa cells express 2000 -3000 TNF-R1 molecules/cell without expressing detectable amounts of TNF-R2 (data not shown). Therefore, TNFinduced TRAF1 transcription in HeLa cells should be mediated exclusively by the TNF-R1.
Next, we investigated whether other receptors that activate NF-B are also capable of inducing transcription of the TRAF1 gene. In fact, stimulation of CD40, IL-1-R, and the TCR also resulted in a strong induction of the TRAF1 gene (Fig. 2, B and  C). In contrast, treatment of HeLa cells with TRAIL-Flag-M2 complex or anti-Apo1 monoclonal antibody, agonists for the receptors Apo1/Fas, DR4 and DR5, respectively, known to be poor activators of NF-B in the HeLa cells studied here, failed to induce TRAF1 (Fig. 2B). Because it has been shown that TRAF1 message is up-regulated in TCR-triggered T-cell hybridomas (19), we investigated whether TNF can induce TRAF1 also in the T-cell line DII/27. As shown in Fig. 2C, both TNF treatment and triggering of the TCR resulted in induction of TRAF1 in this cell line. However, TCR-mediated induction of TRAF1 is not mediated via endogenous production of TNF or LT␣, because high concentrations of the respective neutralizing antibodies showed no effect on TCR-mediated induction of TRAF1 (Fig. 2C). It has been shown that TRAF1 can form heteromers with TRAF2, a molecule that is involved in activation of NF-B and JNK by some of the aforementioned receptors. In addition, there are indications that TRAF1 synergizes with TRAF2 in the CD30-dependent activation of NF-B (4). It is therefore tempting to speculate that TRAF1 functions in modulating TRAF2-dependent responses. The fact that A20, TRIP, and cIAP2, which are all known to regulate TRAF2mediated NF-B activation, associate also with TRAF1 supports this assumption.
Overexpression of an N-terminal Deletion Mutant of TRAF1 Interferes with TNF-induced JNK Activation-To get first insights in a putative regulatory role of TRAF1 in TNF signaling, we stably introduced a FLAG-tagged TRAF1 expression construct in HeLa cells. We obtained two clones, HeLa-TRAF1-a and -b, that significantly overexpress TRAF1 (Fig. 3, A and B). Comparison of the relative levels of mRNA of TRAF1, TRAF2, TRAF4, xIAP, and the housekeeping proteins L32 and GAPDH in Figs. 3B and 2A showed that the amount of TRAF1 message in HeLa-TRAF clones (Fig. 3B) is somewhat higher compared with that in TNF-treated HeLa cells (Fig. 2A). Although the intensity of the TRAF1 band in TNF-treated HeLa cells is significantly lower than that of xIAP and TRAF2 and 4 ( Fig.  2A), in HeLa-TRAF1-a and -b the intensity of the TRAF1 band  (lanes 4, 6, 8, 10, and 12) or 100-fold (lanes 5, 7, 9, 11, and 13) molar excess. As a negative control nonspecific competition was performed by adding an unlabeled double-stranded oligonucleotide encompassing the unrelated Sp1-binding site in 20-fold (lane 14) or 100-fold (lane 15) molar excess. The specific NF-B/Rel complex is indicated. n.s. corresponds to a nonspecific binding. F, HeLa cells were transiently transfected with 10 g of TRAF1 promoter reporter plasmids in which the B sites B1, B3, and B5, respectively, were disrupted. 18 h after transfection, cells were stimulated with TNF (150 units/ml) for an additional 24 h. Relative activity refers to luciferase activity obtained with unstimulated cells transfected with the same plasmids.
is increased relative to the mRNA of these molecules. In addition, the FLAG-tagged TRAF1 gene product could be detected by cytofluorometry using FLAG-specific antibodies (Fig. 3A). We found that both clones exert a sustained activation of JNK after TNF treatment (Fig. 3C). In addition, in all experiments with the TRAF1-overexpressing cells we found a slight but reproducible increase of background JNK activity. Ligand-independent activation of JNK by TRAF proteins, in particular upon transient overexpression, is well documented for other TRAFs (9). To control the specificity of this effect, we also generated a polyclonal population of HeLa transfectants stably overexpressing a deletion mutant of TRAF1 where the N-terminal 184 amino acids had been replaced by the green fluorescent protein (HeLa-GFP-TRAF1(185-408)). After cell sorting for a high expression subpopulation, the cells were analyzed for TNF-dependent JNK activation (Fig. 4). As shown in Fig. 4C, in these cells TNF-induced JNK activation occurs at a significantly reduced level. Nevertheless, we cannot rule out that the effects of TRAF1 and TRAF1(185-408), respectively, on JNK activity are caused by TRAF1-dependent titration of other TRAFs or TRAF-binding proteins involved in regulation of JNK activity.
antibody anti-Apo-1 (50 ng/ml), or TRAIL-Flag-M2 complex (200 ng/ml). Again RNAs were isolated and analyzed with the hApo5 template set in RNase protection assays. C, TCR-mediated induction of TRAF1. DII/27 cells were treated as indicated with TNF (10 ng/ml) or the agonistic TCR-specific monoclonal antibody anti-CD3 (2 g/ml). Stimulation of the TCR was also performed in the presence of neutralizing TNF (30 g/ml) and LT␣ (30 g/ml) antibodies. RNAs were analyzed as described for panels A and B. Flag-tagged TRAF1 was detected by intracellular staining with the anti-FLAG-specific monoclonal antibody M2 and a fluorescein isothiocyanate-conjugated goat anti-mouse secondary antibody. For RNase protection assays 10 g of RNA isolated from untreated HeLa cells and the HeLa transfectants HeLa-TRAF1-a and -b were analyzed with the hApo5B Multi-probe template set for steady state levels of various members of the TRAF and IAP protein families. C, kinetics of TNFinduced JNK activation in clones described in panel A. HeLa, Hela-TRAF1-a, and HeLa-TRAF1-b cells were treated with TNF (5 ng/ml) for the indicated times, and JNK activity was measured by immunocomplex kinase assay with GST-c-Jun(1-79) as substrate.

TRAF1 Modulates TNF-and IL-1-induced Activation of NF-
B-Some recent reports point to a possible costimulatory role of TRAF1 in NF-B activation by the CD30 receptor and the latent infection membrane protein 1 (LMP1) of Epstein-Barr virus (4,34). To further examine the role of the TRAF1 protein in cytokine signaling, HeLa-GFP-TRAF1(185-408) cells were analyzed in respect to TNF-induced NF-B activation. TNFinduced NF-B activation was significantly reduced in HeLa-GFP-TRAF1(185-408) cells as shown by band shift assays (Fig.  5A) as well as reporter plasmid assays (Fig. 5B). The same inhibitory effect on NF-B activation was found in HeLa cells when the reporter plasmid was transiently cotransfected along with the expression vector for GFP-TRAF1(185-408). Expression of full-length TRAF1 augmented TNF-induced NF-B activation (Fig. 5C). Interestingly, IL-1-induced NF-B activation is affected in the same way as TNF-mediated NF-B activation upon expression of GFP-TRAF1(185-408) or full-length TRAF1 (Fig. 5C). These data point to a stimulatory role of TRAF1 in cytokine-induced NF-B-and JNK activation. Interestingly, cIAP2 has a similar function for TNF-induced NF-B activation (42), and its transcription is coordinately regulated with TRAF1 (Fig. 2). Hence, it is possible that the full impact of TRAF1 expression is more obvious in cells overexpressing both TRAF1 and cIAP2. Of course the same holds true for analysis of cIAP2 function.
TRAF1 as well as TRAF2,3,5 and 6 physically interact with the mitogen-activated protein kinase kinase kinase-related serine-threonine kinase NIK (9), which has the capability to phosphorylate the IB kinases ␣ and ␤, thereby linking the members of the TNF receptor superfamily and the IL-1 receptor to the phosphorylation-dependent activation of NF-B (43,44). In contrast to TRAF2, 5, and 6, TRAF1 and TRAF3 are unable to stimulate the catalytic activity of NIK. It has been therefore speculated that TRAF1 may sequester NIK and thereby negatively regulate NF-B activation (9). However, as shown in Fig. 5C, overexpression of TRAF1 in HeLa cells enhances TNF as well as IL-1-induced NF-B activation, arguing for a positive modulatory role of this molecule in NF-B acti- For immunoblotting GFP and GFP-TRAF1(185-408) were immunoprecipitated from cytosolic extracts of 1.5 ϫ 10 6 cells using a rabbit polyclonal IgG fraction against GFP, transferred to nitrocellulose, and detected with a GFP-specific monoclonal antibody and an AP-conjugated secondary antibody. C, GFP-TRAF1(185-408)-expressing cells and control cells were treated with TNF (5 ng/ml) for the indicated times, and JNK activity was measured by immunocomplex kinase assay with GST-c-Jun(1-79) as substrate.
FIG. 5. TRAF1 modulates TNF-mediated NF-B activation. A, electrophoretic mobility shift assay analysis of NF-B activation in HeLa and HeLa-GFP-TRAF1(185-408) cells. 3 ϫ 10 6 cells were treated with the indicated concentrations of TNF for 30 min, and nuclear extracts were prepared and analyzed with a 32 P-labeled NF-B specific oligonucleotide probe. A control reaction was performed with a 100-fold excess of cold competitor oligonucleotide. B, HeLa-GFP (stippled bars) and HeLa-GFP-TRAF1(185-408) (filled bars) cells were transfected with a 3ϫ NF-B-luciferase reporter and ␤-galactosidase standard. One day later, the cultures were stimulated for 6 h with TNF or IL-1␤ or left untreated. Then luciferase activity was determined and normalized against ␤-galactosidase activity. C, HeLa cells were transiently transfected with TRAF1 (hatched bars), GFP-TRAF1(185-408) (filled bars), or a control plasmid (stippled bars) along with a 3ϫ NF-B-luciferase reporter and ␤-galactosidase standard. One day later, the cultures were stimulated for 6 h with TNF or IL-1␤ or left untreated, and subsequently luciferase activity was determined and normalized against ␤-galactosidase activity.
vation. This is also consistent with the fact that TRAF1 overexpression inhibits antigen-induced apoptosis of CD8 ϩ T lymphocytes (36) because this process is driven by TNF via the TNF-R2 (45), and NF-B activation can mediate protection from the cytotoxic effects of TNF (46 -49). In fact, Wang et al. (48), have recently shown that transient overexpression of TRAF1, TRAF2, cIAP1, and cIAP2 together have a pronounced inhibitory effect on TNF-induced apoptosis. We have therefore analyzed the HeLa-TRAF1-GFP cells in respect to TNF-induced apoptosis. As shown in Fig. 6 HeLa-TRAF1-GFP cells are significantly protected against TNF-induced apoptosis 3 h after TNF receptor stimulation. However, this protective effect is transient because HeLa and HeLa-TRAF1-GFP cells are almost comparably sensitive toward the apoptotic action of TNF in overnight assays (data not shown), perhaps because of a limitation in other antiapoptotic factors that act in concert with TRAF1, e.g. TRAF2, cIAP1, or cIAP2.