Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Interferon γ (IFNγ) and Tumor Necrosis Factor α Synergism in ME-180 Cervical Cancer Cell Apoptosis and Necrosis

Interferon γ (IFNγ) and Tumor Necrosis Factor α Synergism in ME-180 Cervical Cancer Cell... THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 16, Issue of April 20, pp. 13153–13159, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Interferon g (IFNg) and Tumor Necrosis Factor a Synergism in ME-180 Cervical Cancer Cell Apoptosis and Necrosis IFNg INHIBITS CYTOPROTECTIVE NF-kB THROUGH STAT1/IRF-1 PATHWAYS* Received for publication, August 22, 2000, and in revised form, December 15, 2000 Published, JBC Papers in Press, January 18, 2001, DOI 10.1074/jbc.M007646200 ¶i i Kyoungho Suk‡§ , Inik Chang‡ , Yun-Hee Kim**, Sunshin Kim**, Ja Young Kim**, Hocheol Kim §, and Myung-Shik Lee‡**‡‡ From the ‡Clinical Research Center, Samsung Biomedical Research Institute and **Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul 135-710, Korea and §Graduate School of East-West Medical Science, Kyunghee University, Seoul, 130-701, Korea We investigated the molecular mechanism of the syn- wide variety of biological activities such as induction of septic ergism between interferon g (IFNg) and tumor necrosis shock, activation of local inflammatory responses, and fever factor a (TNFa) documented in a variety of biological generation as an endogenous pyrogen (1). TNFa also kills var- occasions such as tumor cell death and inflammatory ious tumor cell lines in vitro and mediates anti-tumor effect in responses. IFNg/TNFa synergistically induced apoptosis vivo (2). TNFa exerts its biological effects by binding to two of ME-180 cervical cancer cells. IFNg induced STAT1 types of cell surface receptors with molecular masses of 55 kDa phosphorylation and interferon regulatory factor 1 (p55) and 75 kDa (p75). TNFa cytotoxicity is mostly mediated (IRF-1) expression. Transfection of phosphorylation-de- by p55 receptors (3). After the ligation of p55 receptors, a fective STAT1 inhibited IFNg/TNFa-induced apoptosis, canonical apoptotic signal transduction pathway is initiated. whereas IRF-1 transfection induced susceptibility to The cytoplasmic death domain of p55 receptor interacts with TNFa. Dominant-negative IkBa transfection sensitized the death domain of intracellular adapter molecules such as ME-180 cells to TNFa. IFNg pretreatment attenuated TRADD (TNF receptor-associated death domain protein) and TNFa- or p65-induced NF-kB reporter activity, whereas FADD (Fas-associated death domain protein), which leads to it did not inhibit p65 translocation or DNA binding of the activation of initiator caspases (4). This, in turn, triggers NF-kB. IRF-1 transfection alone inhibited TNFa-in- the caspase cascade and ultimately results in apoptotic cell duced NF-kB activity, which was reversed by coactiva- death. tor p300 overexpression. Caspases were activated by In many cases, the anti-tumor effect of TNFa was enhanced IFNg/TNFa combination; however, caspase inhibition by IFNg (5) or metabolic inhibitors such as cycloheximide and did not abrogate IFNg/TNFa-induced cell death. In- actinomycin D (6). Although these metabolic inhibitors are stead, caspase inhibitors directed IFNg/TNFa-treated believed to block the synthesis of cytoprotective proteins, the ME-180 cells to undergo necrosis, as demonstrated by effects of IFNg might be mediated by the induction of new Hoechst 33258/propidium iodide staining and electron microscopy. Taken together, our results indicate that proteins that increase the sensitivity of target cells to TNFa. IFNg and TNFa synergistically act to destroy ME-180 IFNg/TNFa synergism also has been reported in biological tumor cells by either apoptosis or necrosis, depending responses other than tumor cell killing. For instance, the two on caspase activation, and STAT1/IRF-1 pathways initi- cytokines synergistically up-regulated the expression of nu- ated by IFNg play a critical role in IFNg/TNFa syner- merous genes, including ICAM-1 (intercellular adhesion mole- gism by inhibiting cytoprotective NF-kB. IFNg/TNFa cule 1), IP-10, and major histocompatibility complex class I synergism appears to activate cell death machinery in- heavy chain (7–9). However, the molecular mechanism of the dependently of caspase activation, and caspase activa- synergism between the two cytokines is not clearly understood. tion seems to merely determine the mode of cell death. It has been reported that IFNg increases the expression of TNFa receptors (10). However, because the sensitivity of the cells to TNFa is not simply correlated with the level of TNFa The pleiotropic proinflammatory cytokine TNFa exerts a receptor expression (11, 12), up-regulation of TNFa receptor alone does not adequately explain the cytokine synergism in the anti-tumor action. * This work was supported by National Research Laboratory Grants In the current work, we utilized ME-180 human cervical 2000-N-NL-01-C-232 from the Korea Institute of Science and Technol- cancer cells to investigate the molecular mechanism of syner- ogy Evaluation and Planning and by Science Research Center Grants from Korea Science and Engineering Foundation. The costs of publica- gistic anti-tumor effects of IFNg/TNFa. We also studied the tion of this article were defrayed in part by the payment of page role of caspase activation in ME-180 cell death by IFNg/TNFa charges. This article must therefore be hereby marked “advertisement” synergism. Our results indicate that 1) IRF-1 induction after in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. STAT1 activation by IFNg plays a central role in synergistic Supported by Brain Korea 21 project from the Ministry of Educa- tion, Korea. These authors contributed equally to this work. ‡‡ Recipient of Juvenile Diabetes Foundation International Research benzyloxycarbonyl-Val-Ala-Asp(OCH )-CH -fluoromethyl ketone; BD- 3 2 Grant 1-1999-760). To whom correspondence should be addressed. Tel.: fmk, t-butoxycarbonyl-Asp(OCH )-CH -fluoromethyl ketone; z-DEVD- 3 2 82-2-3410-3436; Fax: 82-2-3410-3849; E-mail: mslee@smc.samsung. fmk, benzyloxycarbonyl-Asp(OCH )-Glu(OCH )-Val-Asp(OCH )-CH - 3 3 3 2 co.kr. fluoromethyl ketone; z-IETD-fmk, benzyloxycarbonyl-Ile-Glu(OCH )- The abbreviations used are: TNF, tumor necrosis factor; IFN, int- Thr-Asp(Ome)-CH -fluoromethyl ketone; MTT, 3-[4,5-dimethylthiazol- erferon; IRF, interferon regulatory factor; STAT, signal transducer and 2-yl]-2,5-diphenyltetrazolium bromide; Ac, acetyl; AMC, amidome- activator of transcription; PI, propidium iodide; z-VAD-fmk, thylcoumarin. This paper is available on line at http://www.jbc.org 13153 This is an Open Access article under the CC BY license. 13154 IFNg/TNFa Synergism in ME-180 Cell Death anti-human phospho-STAT1, New England Biolabs) and horseradish tumor cell death by IFNg/TNFa,2)IFNg-induced IRF-1 inhib- peroxidase-conjugated secondary antibodies (anti-rabbit IgG, Amer- its cytoprotective NF-kB transactivation, 3) IFNg/TNFa in- sham Pharmacia Biotech), followed by ECL detection (Amersham Phar- duces ME-180 cell death regardless of caspase activation, and macia Biotech). caspase activation dictates only the mode of cell death between Transient Transfection—ME-180 cells in 6-well plates were co-trans- apoptosis and necrosis. fected with 1 mg of human STAT1 cDNA, dominant-negative mutant STAT1 cDNA (kindly provided by Dr. Hirano, Osaka University, Ja- EXPERIMENTAL PROCEDURES pan), human IRF-1 cDNA (kindly provided by Dr. Taniguchi, Univer- Cell Line and Reagents—ME-180 cervical cancer cell line was ob- sity of Tokyo), or phosphorylation-defective dominant-negative mutant tained from ATCC (Manassas, VA) and grown in Dulbecco’s modified IkBa (14) together with 0.2 mgof lacZ gene (pCH110, Amersham Phar- Eagle’s medium containing 10% fetal bovine serum, 2 mM glutamine, macia Biotech) using LipofectAMINE reagent (Life Technologies, Inc.). and penicillin-streptomycin (Life Technologies, Inc.). Recombinant hu- 48 h after the transfection, cells were treated with cytokines. After man IFNg was purchased from R&D Systems (Minneapolis, MN). Re- another 48 h, the cells were fixed with 0.5% glutaraldehyde for 10 min combinant human TNFa was generously provided by Dr. T. H. Lee at room temperature and stained with X-gal (5-bromo-4-chloro-3-indo- (Yonsei University, Seoul, Korea). Caspase inhibitors (z-VAD-fmk, ben- lyl b-D-galactopyranoside; 1 mg/ml) in 4 mM potassium ferricyanide, 4 zyloxycarbonyl-Val-Ala-Asp(OCH )-CH -fluoromethyl ketone; BD-fmk, mM potassium ferrocyanide, 2 mM magnesium chloride at 37 °C for 3 2 t-butoxycarbonyl-Asp(OCH )-CH F; z-DEVD-fmk, benzyloxycarbonyl- detection of blue cells. At least 200 blue cells were counted for each 3 2 Asp(OCH )-Glu(OCH )-Val-Asp(OCH )-CH -fluoromethyl ketone; z-IETD- experiment, and transfection efficiency was 10 –35%. Results were pre- 3 3 3 2 fmk, benzyloxycarbonyl-Ile-Glu(OCH )-Thr-Asp(OCH )-CH -fluoromethyl sented as means 6 S.E. (n 5 3). 3 3 2 ketone) were purchased from Enzyme Systems (Livermore CA), and cathep- NF-kB Reporter Assays—NF-kB reporter activity was measured us- sin B inhibitor FA (benzyloxycarbonyl-Phe-Ala-CH -fluoromethyl ketone) ing the dual-luciferase reporter assay system (Promega, Madison, WI). and MG-132 (carbobenzoxyl-leucinyl-leucinyl-leucinal-H, also called Z-LLL) In brief, ME-180 cells in 12-well plates were co-transfected with 0.5 mg were from Calbiochem. All other chemicals were obtained from Sigma, of NF-kB-responsive reporter gene construct carrying two copies of kB unless stated otherwise. sequences linked to luciferase gene (IgGk NF-kB-luciferase, generously Assessment of Cytotoxicity by 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphe- provided by Dr. G. D. Rosen, Stanford University, Stanford, CA) (15) nyltetrazolium Bromide (MTT) Assay—Cells (3 3 10 /well) were seeded together with 0.1 mgof Renilla luciferase gene under hamster sarcoma in 96-well plates and treated with various combinations of cytokines for virus thymidine kinase promoter (pRL-TK, Promega) using Lipo- the indicated time periods. The optimal concentrations of the cytokines fectAMINE reagent (Life Technologies, Inc.). 24 h after the transfection, for the cytotoxic action were 100 units/ml for IFNg and 10 ng/ml for cells were treated with cytokines. After 5 h, activities of firefly lucifer- TNFa. In some experiments, cells were pretreated with caspase inhib- ase and Renilla luciferase in transfected cells were measured sequen- itors or MG-132 for 1 h before cytokine treatment. After cytokine tially from a single sample using the dual-luciferase reporter assay treatment, the medium was removed, and MTT (0.5 mg/ml) was added, system (Promega). Results were presented as firefly luciferase activity followed by incubation at 37 °C for2hinCO incubator. After a brief normalized to Renilla luciferase activity. In some experiments, cells centrifugation, supernatants were carefully removed, and Me SO was were co-transfected before cytokine treatment with NF-kB p65 (16) or added. After insoluble crystals were completely dissolved, absorbance coactivator p300 expression plasmid (0.5 mg; kindly provided by Dr. at 540 nm was measured using a Thermomax microplate reader (Mo- Livingston, Harvard Medical School, Boston, MA) (17) along with NF- lecular Devices). Results were presented as means 6 S.E. (n 5 3). kB-responsive reporter plasmid (0.5 mg) and pRL-TK (0.1 mg). Results Morphological Analysis of Apoptotic Cells—Morphological changes in were presented as means 6 S.E. (n 5 3). the nuclear chromatin of cells undergoing apoptosis were detected by Immunofluorescence Staining—ME-180 cells seeded onto chamber staining with 2.5 mg/ml bisbenzimide Hoechst 33258 fluorochrome (Cal- slides (Lab-Tek, Nalge Nunc International, Naperville, IL) were fixed in biochem), followed by examination on a fluorescence microscope. In 4% paraformaldehyde for 30 min at room temperature and then in cold some experiments, cytokine-treated cells were double-stained with pro- methanol for 10 min at 220 °C. Fixed cells were permeabilized in 0.1% pidium iodide (PI, 2.5 mg/ml) and Hoechst 33258 (2.5 mg/ml) to distin- Triton X-100, 0.1% sodium citrate for 3 min at 4 °C and then sequen- guish apoptotic cells from necrotic cells. Intact blue nuclei, condensed/ tially incubated with mouse anti-p65 antibody (Santa Cruz Biotechnol- fragmented blue nuclei, condensed/fragmented pink nuclei, and intact ogy, Santa Cruz, CA), biotinylated anti-mouse IgG, and streptavidin- pink nuclei were considered viable, early apoptotic, late apoptotic, and fluorescein isothiocyanate. Stained cells were examined on a necrotic cells, respectively (13). Transmission electron microscopy was fluorescent microscope. carried out essentially as previously described (13). In brief, cells were Electrophoretic Mobility Shift Assay—Nuclear extracts were pre- fixed in 4% glutaraldehyde, 1% paraformaldehyde, 0.2 M phosphate, pH pared from ME-180 cells treated with cytokines as previously described 7.2, at 4 °C for 2 h. After two washes in 0.2 M phosphate, the cell pellet (18). Synthetic double-strand oligonucleotides of consensus NF-kB was post-fixed with 2% OsO in the same buffer for 30 min. The pellet binding sequence, GAT CCC AAC GGC AGG GGA (Promega), were was dehydrated in ethanol and then in 100% propylene oxide, followed end-labeled with [g- P]ATP using T4 polynucleotide kinase. Nuclear by embedding overnight at 37 °C for another 3 days at 60 °C. Ultrafine extract was incubated with the labeled probe in the presence of poly- sections were cut and examined on an electron microscope (Hitachi (dI-dC) in a binding buffer containing 20 mM HEPES at room temper- H7100, 75 kV). ature for 30 min. For supershift assays, a total of 0.2 mg of antibodies DNA Ploidy Analysis—Cells were suspended in phosphate-buffered against p65 or p50 subunit of NF-kB were included in the reaction. saline, 5 mM EDTA and fixed by adding 100% ethanol dropwise. RNase DNA-protein complexes were resolved by electrophoresis in a 5% nonde- A (40 mg/ml) was added to resuspended cells, and the incubation was naturing polyacrylamide gel, dried, and visualized by autoradiography. carried out at room temperature for 30 min. PI (50 mg/ml) was then added for flow cytometric analyses. RESULTS Assessment of Caspase Activity—Caspase-3- or -8-like activity was IFNg and TNFa Synergistically Induced the Apoptosis of measured using a caspase assay kit (Pharmingen, San Diego, CA) ME-180 Cells—First we screened several tumor cell lines to according to the supplier’s instruction. In brief, caspase-3 or -8 fluoro- genic substrates (Ac-DEVD-AMC or Ac-IETD-AMC) were incubated assess their sensitivity to IFNg/TNFa-induced cytotoxicity with cytokine-treated cell lysates for1hat37 °C, then AMC liberated (data not shown). Cytotoxic synergism between IFNg and from Ac-DEVD-AMC or Ac-IETD-AMC was measured using a fluoro- TNFa was most evident in ME-180 cells. Although either cy- metric plate reader with an excitation wavelength of 380 nm and an tokine alone exhibited no significant cytotoxicity, the combina- emission wavelength of 420- 460 nm. tion of the two cytokines significantly reduced ME-180 cell Western Blot Analysis—Cells were lysed in triple-detergent lysis viability (Fig. 1A). The cytokine cytotoxicity was dependent on buffer (50 mM Tris-Cl, pH 8.0, 150 mM NaCl, 0.02% sodium azide, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 1 mM phenylmeth- the dose of IFNg used. However, concentration higher than 100 ylsulfonyl fluoride). Protein concentration in cell lysates was deter- units/ml did not further increase the cytotoxicity (Table I). The mined using the Bio-Rad protein assay kit. An equal amount of protein reduction of cell viability was due to apoptosis as demonstrated for each sample was separated by 10 or 12% SDS-polyacrylamide gel by Hoechst 33258 staining and DNA ploidy analysis. IFNg/ electrophoresis and transferred to Hybond ECL nitrocellulose mem- TNFa treatment induced nuclear condensation and fragmen- branes (Amersham Pharmacia Biotech). The membranes were blocked tation (Fig. 1B) and led to the appearance of sub-diploid cells with 5% skim milk and sequentially incubated with primary antibodies (rabbit anti-human IRF-1, Santa Cruz; rabbit anti-human STAT1 and (Fig. 1C), which are hallmarks of apoptotic cells. DNA ploidy IFNg/TNFa Synergism in ME-180 Cell Death 13155 TABLE II Cytotoxic effects of sequential treatment of cytokines a b Treatment % Viability None 100 IFNg 1 TNFa (48 h) 22.1 6 3.1 IFNg (48 h) 98.2 6 3.4 IFNg (24 h) and then TNFa (48 h) 52.3 6 2.4 TNFa (48 h) 101.5 6 2.9 TNFa (24 h) and then IFNg (48 h) 90.3 6 3.3 ME-180 cells were treated with cytokines as indicated, either simul- taneously or sequentially. Treatment with IFNg (100 units/ml) for 24 h followed by TNFa (10 ng/ml) treatment for 48 h induced a significant cytotoxicity. However, sequential treatment with the two cytokines in a reverse order did not significantly affect ME-180 cell viability, indicat- ing the priming role of IFNg in TNFa-induced ME-180 cell death. Cell viability was assessed by MTT assays. Viability of untreated cells was set to 100%. Results are means 6 S.E. (n 5 3). FIG.2. IFNg activates STAT1 (A) and induces IRF-1 expression (B) in ME-180 cells. Western blot analyses demonstrated that treat- FIG.1. IFNg/TNFa synergistically induces ME-180 cell apopto- ment of ME-180 cells with IFNg induced STAT1 expression (24-h treat- sis. A combination of IFNg (100 units/ml) and TNFa (10 ng/ml), but not ment) as well as its phosphorylation (30-min treatment) (A). IFNg also either cytokine alone, induced ME-180 cell death. Cell viability was induced IRF-1 expression at 24 h after treatment, and the expression assessed by MTT assays after treatment with the cytokines for 48 h (A). was further increased at 48 h after the treatment (B). However, TNFa Induction of ME-180 cell death was due to apoptosis, as demonstrated alone did not change the expression of either STAT1 or IRF-1. C, by chromatin condensation in Hoechst 33258 staining (B) or the ap- untreated control; I, IFNg (100 units/ml); T, TNFa (10 ng/ml); D, IFNg pearance of sub-G peak in flow cytometric analyses (C) at 24 h after plus TNFa. cytokine treatment. with IFNg did not have the same effects, indicating that IFNg TABLE I confers susceptibility to TNFa on ME-180 cells through induc- Dose response of cytokine cytotoxicity tion or up-regulation of certain genes in ME-180 cells. Because Treatments % Viability STAT1 and IRF-1 are known to be canonical intracellular sig- None 100 nal-transducing molecules in IFNg signaling, we investigated IFNg (1 unit/ml) 1 TNFa 95.6 6 3.6 the involvement of STAT1/IRF-1-signaling pathways in IFNg/ IFNg (10 units/ml) 1 TNFa 76.2 6 2.5 TNFa synergism on ME-180 cell apoptosis. IFNg, but not IFNg (100 units/ml) 1 TNFa 23.4 6 3.1 TNFa, induced phosphorylation of STAT1 and up-regulated IFNg (1000 units/ml) 1 TNFa 24.7 6 2.8 IRF-1 expression in ME-180 cells (Fig. 2). Furthermore, the ME-180 cells were treated with increasing concentrations of IFNg transfection of phosphorylation-defective dominant-negative and TNFa (10 ng/ml) for 48 h, and then cell viability was assessed by MTT assays. Viability of untreated cells was set to 100%. Results are mutant of STAT1 significantly inhibited IFNg/TNFa-induced the means 6 S.E. (n 5 3). IFNg alone at all concentrations tested did ME-180 cell death, indicating that IFNg-induced STAT1 acti- not have a significant cytotoxicity (data not shown). One hundred vation is critical for the induction of TNFa susceptibility (Fig. units/ml IFNg and 10 ng/ml TNFa were the optimal concentrations for 3A). We next asked whether IRF-1, a downstream mediator of cytotoxic assays. STAT1, is responsible for the priming effects of IFNg. Trans- fection of IRF-1 conferred TNFa susceptibility on ME-180 cells in a dose-dependent manner, indicating a central role for IRF-1 assays also indicated that the effect of IFNg/TNFa was not due in the sensitization of ME-180 cells to TNFa-induced apoptosis to the growth arrest as was shown by the absence of decrease in (Fig. 3, B and C). the percentage of cells in the S phase. Inhibition of Cytoprotective NF-kB Activity by IFNg—TNFa is IFNg/TNFa Synergism Involved IFNg-induced STAT1 Acti- known to initiate both death and survival signals, and recent stud- vation and IRF-1 Induction—Based on our results that the ies on TNFa-induced survival signal suggested an important role of combination of IFNg and TNFa, but not either cytokine alone, NF-kB activation (19 –22). Thus, we investigated how IFNg in- induced ME-180 cell death, we explored the possibility that duces susceptibility to TNFa-induced cytotoxicity by examining the IFNg sensitizes ME-180 cells to TNFa-mediated cytotoxicity. role of NF-kB in ME-180 cell death and its possible regulation by This was first tested by sequential treatment of ME-180 cells IFNg. Treatment of ME-180 cells with a proteasome inhibitor (MG- with the two cytokines. After IFNg treatment, TNFa alone was 132), which is known to inhibit NF-kB activation (23), rendered the sufficient to induce a significant cytotoxicity in ME-180 cells cells sensitive to TNFa-induced apoptosis (Fig. 4A), suggesting the (Table II). However, sequential treatment with TNFa and then cytoprotective role of NF-kB. Also, upon the transfection of phos- 13156 IFNg/TNFa Synergism in ME-180 Cell Death FIG.3. A key role for STAT1/IRF-1 signaling in IFNg/TNFa synergism. A, transient transfection of phosphorylation- defective STAT1 dominant-negative mu- tant (DN STAT1) significantly inhibited IFNg/TNFa cytotoxicity, as demonstrated by counting blue cells co-expressing lacZ at 48 h after cytokine treatment (IFNg, 100 units/ml; TNFa, 10 ng/ml). B, trans- fection of IRF-1 cDNA (1 mg) induced sus- ceptibility to TNFa. In contrast to empty vector (pcDNA3) transfectants, treatment of IRF-1 transfectants with TNFa alone for 48 h significantly decreased the num- ber of blue cells. C, the effects of IRF-1 were dependent upon the dose of IRF-1 cDNA (0.1, 0.5, 1, and 2 mg) used in the transient transfection. The number of blue cells upon transfection with an empty vector without TNFa treatment was set to 100%. FIG.4. Inhibition of cytoprotective NF-kBbyIFNg. A, inhibition of NF-kB by proteasome inhibitor MG-132 sensi- tized ME-180 cells to TNFa. ME-180 cells were treated with either MG-132 (0.5 mM) alone or in combination with TNFa (10 ng/ml) for 48 h, and then cell viability was assessed by MTT assays. B, inhibition of NF-kB by transfection of dominant-nega- tive mutant IkBa (DN IkBa) also ren- dered ME-180 cells sensitive to TNFa treatment. Viability of ME-180 cells co- transfected with dominant-negative IkBa and lacZ was significantly decreased by TNFa treatment (24 h), in contrast to the cells co-transfected with an empty vector (pcDNA3) and lacZ. C and D, NF-kB re- porter assays revealed that pretreatment (24 h, 100 units/ml) of ME-180 cells with IFNg inhibited TNFa-induced NF-kB ac- tivity (C). IFNg treatment (48 h) also in- hibited NF-kB reporter activity induced by p65 transfection (NF-kB p65)(D). Transiently transfected cells were treated with cytokines for the indicated time period before NF-kB reporter assays (C and D). phorylation-defective dominant-negative mutant IkBa, TNFa of IFNg on NF-kB. Transfection of IRF-1 alone was sufficient to alone induced a significant cytotoxicity, further supporting the cy- inhibit TNFa-induced NF-kB activity, indicating a central role of toprotective role of NF-kB (Fig. 4B). NF-kB reporter assays indi- IRF-1 in the inhibition of NK-kB transactivation by IFNg (Fig. 7A). cated that IFNg pretreatment attenuated TNFa-induced NF-kB We also investigated the possible mechanism of interference be- activity, suggesting that IFNg synergizes with TNFa for ME-180 tween IRF-1 and NF-kB. Transfection of p300 coactivator abro- cell apoptosis by inhibiting TNFa-induced cytoprotective NF-kB gated the inhibitory effect of IFNg treatment (Fig. 7B) or IRF-1 activity (Fig. 4C). IFNg pretreatment, however, did not inhibit transfection (Fig. 7C)onTNFa-induced NF-kB activity, suggesting nuclear translocation of p65 (Fig. 5) or DNA binding of NF-kB the possibility of coactivator competition between IFNg-induced induced by TNFa treatment (Fig. 6). Also, IFNg did not inhibit IRF-1 and TNFa-induced NF-kB. TNFa-induced degradation of IkBa (data not shown). However, Inhibition of Caspases Directed ME-180 Cells to Undergo IFNg treatment did inhibit the NF-kB reporter activity induced by Necrotic Cell Death—We next investigated whether the activa- transfection of p65 subunit of NF-kB (Fig. 4D), indicating that tion of caspases is involved in the IFNg/TNFa-induced apopto- IFNg directly inhibited NF-kB-mediated transactivation within the sis of ME-180 cells. Cytokine-induced apoptosis of ME-180 cells nuclei without affecting the nuclear translocation or DNA binding was accompanied by the activation of caspase-3-like activity, as of NF-kB. We next studied if IRF-1 mediates this inhibitory action demonstrated by the cleavage of Ac-DEVD-AMC in IFNg/ IFNg/TNFa Synergism in ME-180 Cell Death 13157 TNFa-treated cells (Fig. 8). Cytokine treatment also induced ergism in ME-180 cell apoptosis. However, dominant-negative the cleavage of Ac-IETD-AMC, indicating concurrent activation STAT1 did not completely abolish cytotoxicity by IFNg/TNFa, of caspase-8-like activity (data not shown). However, pretreat- and IRF-1 transfection could not be completely substituted for ment with broad-spectrum caspase inhibitors such as z-VAD- IFNg. IFNg induces STAT1 as well as IRF-1, and some cellular fmk or BD-fmk failed to inhibit ME-180 cell death by IFNg/ responses to IFNg are reported to be mediated by both STAT1 TNFa synergism despite the activation of multiple caspases and IRF-1 (24 –25). Neither STAT1 or IRF-1 alone may not (Fig. 9A). Instead, IFNg/TNFa in the presence of caspase in- explain all of the priming effect of IFNg in TNFa-induced hibitors unexpectedly induced the necrosis of ME-180 cells, as death. The role of IRF-1 in the induction of apoptosis by DNA judged by the swelling of dying cells on a light microscope (data damage or IFNg has been previously suggested (26 –28), which not shown). Hoechst 33258/PI staining and electron microscopy supports the proapoptotic action of IRF-1. Previous work in our confirmed the necrosis of the cells (Fig. 9, B and C). To study laboratory also showed that IRF-1 plays a central role in IFNg/ the involvement of individual caspases in the switching process TNFa-induced apoptosis of pancreatic islet b-cells in autoim- from apoptosis to necrosis, cells were pretreated with inhibitors mune diabetes. Caspase induction has been suggested as a specific for individual caspases instead of z-VAD-fmk. Because possible downstream event after IRF-1 induction in IFNg-in- we observed the activation of caspase-3 and -8 in the cytokine- duced apoptosis (28). Although RNase protection assays re- treated ME-180 cells, we tested the effects of z-DEVD-fmk and z-IETD-fmk alone or in combination. The z-DEVD-fmk and z-IETD-fmk acted additively in conversion from apoptosis to K. Suk, I. Chang, Y.-H. Kim, S. Kim, J. Y. Kim, and M.-S. Lee, submitted for publication. necrosis, suggesting the involvement of multiple caspases in determining the mode of ME-180 cell death (Table III). DISCUSSION Here we present evidence that STAT1/IRF-1 pathways ini- tiated by IFNg play a central role in IFNg/TNFa synergism in the induction of ME-180 cell apoptosis. Transfection of domi- nant-negative STAT1 abolished IFNg/TNFa synergism, whereas transfection of IRF-1 sensitized ME-180 cells to TNFa-induced apoptosis. Thus, STAT1 activation and IRF-1 induction by IFNg appear to be important in IFNg/TNFa syn- FIG.6. No significant effects of IFNg on DNA binding of NF-kB protein. A, IFNg pretreatment (100 units/ml, 24 h) did not signifi- cantly affect TNFa-induced kB sequence binding of NF-kB proteins (lanes 4 and 5). The identity of DNA-complexed proteins was confirmed by supershift assays using antibodies (Ab) against p65 (lane 6), p50 (lane 7), or both (lane 8). B, ME-180 cells were similarly treated with increasing doses of IFNg and TNFa as indicated, and then NF-kB was detected by electrophoretic mobility shift assay. IFNg at all concentra- FIG.5. No effects of IFNg on nuclear translocation of p65 sub- unit of NF-kB. As compared with untreated control (A), TNFa treat- tions tested did not significantly influence TNFa-induced DNA binding of NF-kB, indicating that the inability of IFNg to inhibit TNFa-induced ment (45 min, 10 ng/ml) induced nuclear translocation of p65 (B), which was not affected by IFNg pretreatment (24 h, 100 units/ml) (C). DNA binding of NF-kB was not due to the low dose of IFNg used. FIG.7. IRF-1 mediates NF-kB-inhibiting effects of IFNg probably through coactivator competition. A, transfection of IRF-1 inhibited TNFa-induced NF-kB reporter activity in a manner similar to IFNg pretreatment. B, transfection of coactivator p300 abrogated IFNg-mediated inhibition of NF-kB reporter activity. Transiently transfected cells were treated with cytokines for an indicated time period before NF-kB reporter assays (IFNg, 100 units/ml; TNFa, 10 ng/ml). C, ME-180 cells were co-transfected with NF-kB reporter construct and the indicated plasmids, and then the luciferase activity was measured after 24 h. Co-transfection of coactivator p300 also abolished the IRF-1 transfection-mediated inhibition of NF-kB reporter activity. CBP, cAMP-response element-binding protein (CREB)-binding protein. 13158 IFNg/TNFa Synergism in ME-180 Cell Death FIG.8. Activation of caspase-3-like activity by IFNg/TNFa in ME-180 cells. IFNg/TNFa treatment (IFNg, 100 units/ml; TNFa,10 ng/ml) induced cleavage of DEVD-AMC, indicating activation of caspase-3-like activity. Pretreatment of ME-180 cells with z-VAD-fmk before cytokine treatment completely inhibited the caspase activity (IFNg, 100 units/ml; TNFa, 10 ng/ml). vealed that the expression of caspase-1 and -4 was up-regu- lated by IFNg treatment in ME-180 cells (data not shown), there remains yet to be determined how the increases in the expression of these caspases mediate IRF-1 action. In IFNg/ TNFa-induced death of ME-180 cells, caspases seem to be involved in determining the mode of cell death rather than decision between death and survival (see below). Although further works are necessary to completely delineate the downstream signaling pathways after STAT1/IRF-1 in IFNg/ TNFa cytotoxic synergism, our current work indicates that NF-kB is one of the targets of STAT1/IRF-1 action. We demonstrated that IFNg attenuated TNFa-induced NF-kB reporter activity in ME- 180 cells. Also, the inhibition of NF-kB either by transfection of dominant-negative IkB “super repressor” or by treatment with a proteasome inhibitor (MG-132) rendered ME-180 cells sensitive to TNFa-induced apoptosis. These results indicate that IFNg sensi- tizes ME-180 cells to TNFa-induced apoptosis by inhibiting NF-kB- mediated activation of survival signals. Furthermore, this action of IFNg was mediated by IRF-1. It has been previously reported that FIG.9. Induction of necrotic death by IFNg/TNFa in the pres- IRF-1 and NF-kB interact in vitro as well as in vivo for the coop- ence of caspase inhibitors. A, pretreatment of ME-180 cells with erative induction of inflammatory genes (29 –31). In ME-180 cells, broad spectrum caspase inhibitors such as z-VAD-fmk or BD-fmk did however, IRF-1 negatively influenced NF-kB activity. IRF-1 does not block the cytokine-induced cytotoxicity as measured by MTT assays at 48 h after the treatment (IFNg, 100 units/ml; TNFa, 10 ng/ml). B and not seem to directly interact with NF-kB because NF-kB transcrip- C, pretreatment with z-VAD-fmk switched the mode of cell death from tional activity was assessed using a reporter construct containing a apoptosis to necrosis as judged by Hoechst 33258/PI double-staining (B) kB element but not an IRF-1 response element. Thus, in ME-180 and electron microscopy (C). In Hoechst 33258/PI double-staining, cells cells, it is likely that IRF-1 indirectly affects NF-kB transcriptional with blue intact nuclei were viable cells, whereas those with blue fragmented nuclei were early apoptotic cells. Cells with pink intact activity through the regulation of other factors modulating the nuclei were necrotic cells, whereas cells with pink fragmented nuclei transcriptional activity. We also demonstrated that IFNg did not were late apoptotic cells. The values in the parentheses below the block the TNFa-induced translocation of p65 from cytosol to nu- photographs represent the percentage of apoptotic (early or late) or cleus or DNA binding of NF-kB but yet inhibited NF-kB reporter necrotic cells out of the total 500 cells counted (B). activity. These results suggest that IFNg-induced IRF-1 inhibits the nuclear events of NF-kB transactivation but not cytosolic ing the expression of multiple genes involved in the inflammatory events. Our work also showed that transfection of transcriptional responses (34). In sharp contrast, however, our work disclosed that coactivator p300 abolished the inhibition of NF-kB reporter activity IFNg inhibited TNFa-induced NF-kB in ME-180 cells. This novel by IFNg. Transcriptional activation by NF-kB requires multiple signaling pathway of synergism between IFNg/TNFa involving coactivators (32). It has been recently reported that the intracellu- competition between IRF-1 and NF-kB for p300 coactivator may lar amount of the coactivator p300 is limited compared with other not be generalized to other cell types, considering previous reports transcriptional factors and that competition for p300 may regulate showing different signaling patterns in response to IFNg/TNFa transcriptional activity (33). Thus, it is possible that IFNg-induced (34). The same stimulus seems to activate distinct signaling path- IRF-1 competes with TNFa-induced NF-kB for the common coac- ways depending on the cell types. Because of this discrepancy in tivator(s) such as p300, and this competition may be responsible for signal transduction pathways, the final outcome would be different the inhibition of NF-kB transactivation. Then what are the target among different cell types. Some cells would undergo death by genes that are induced by NF-kB and are subject to the inhibitory IFNg/TNFa, whereas other cells may be activated by IFNg/TNFa action of IRF-1? Recently, a role of TNF receptor-associated factor to participate in inflammatory responses. 1 (TRAF2), TRAF2, c-IAP1 (inhibitor of apoptosis (IAP)) and cIAP2 Because IFNa is also known to activate the STAT1-signaling was reported in anti-apoptosis mediated by NF-kB (19). These are pathway, we investigated whether IFNa also synergizes with possible candidates for such target genes. Another puzzling point is TNFa to destroy ME-180 cells. Our results indicated that IFNa what determines how IFNg acts on NF-kB. Previously, IFNg has and TNFa synergistically induced ME-180 cell death, and this been shown to increase TNFa-induced NF-kB activation in enhanc- was accompanied by the activation of STAT1 and inhibition of IFNg/TNFa Synergism in ME-180 Cell Death 13159 TABLE III IFNg synergized with TNFa for apoptosis induction by activat- Effects of various caspase inhibitors on the conversion of ME-180 cell ing STAT1/IRF-1 pathway. We also present evidence that death from apoptosis to necrosis NF-kB activation is a survival signal in TNFa-treated ME-180 a % Viable % Necrotic % Apoptotic Treatment cells, and IFNg inhibits this survival mechanism, resulting in b b b cells cells cells synergistic cytotoxicity with TNFa. Moreover, the mode of ME- None 94.7 3.1 2.2 180 cell death by IFNg/TNFa synergism was dictated by IFNg 1 TNFa 14.7 8.6 76.7 caspase activation. The novel mechanism of IFNg/TNFa syn- IFNg 1 TNFa 1 z-VAD 16.5 78.5 5.0 ergism presented here may also be applicable to other circum- IFNg 1 TNFa 1 DEVD 16.8 25.2 58.0 IFNg 1 TNFa 1 IETD 20.4 34.5 45.1 stances, where a similar cytokine synergism could be found IFNg 1 TNFa 1 DEVD 1 IETD 27.6 55.2 17.2 such as autoimmune destruction of self tissues by cytokines. IFNg 1 TNFa 1 FA 11.2 10.5 78.3 Acknowledgments—We thank Drs. Kye Young Lee, Tae H. Lee, Jae ME-180 cells were pretreated with caspase inhibitors indicated (100 W. Lee, Minho Shong, Soo Young Lee, Il-Seon Park, and Young S. Ahn mM) for 1 h before cytokine treatment for 48 h (IFNg, 100 units/ml; for insightful discussions and technical help. TNFa, 10 ng/ml). Cathepsin B inhibitor, FA, benzyloxycarbonyl-Phe- Ala-CH -fluoromethyl ketone (FA) was used as a negative control. Percentage of viable, necrotic, or apoptotic cells was assessed by REFERENCES double-staining with Hoechst 33258 and PI. 1. Fiers, W. (1991) FEBS Lett. 285, 199 –212 2. Beyaert, R., and Fiers, W. (1994) FEBS Lett. 340, 9 –16 NF-kB reporter activity by IFNa in a manner similar to IFNg. 3. Schulze-Osthoff, K., Ferrari, D., Los, M., Wesselborg, S., and Peter, M. E. (1998) Eur. J. Biochem. 254, 439 – 459 Thus, the cytotoxic priming role of IFNg in IFNg/TNFa syner- 4. Hsu, H., Shu, H. B., Pan, M. G., and Goeddel, D. V. (1996) Cell 84, 299 –308 gism presented in the current study does not seem to be re- 5. Sugarman, B. J., Aggarwal, B. B., Hass, P. E., Figari, I. S., Palladino, M. A., Jr., and Shepard, H. M. (1985) Science 230, 943–945 stricted to IFNg. Rather, the STAT1/IRF-1-signaling pathway 6. Kirstein, M., Fiers, W., and Baglioni, C. (1986) J. Immunol. 137, 2277–2280 that can be initiated by either type I or type II interferon 7. Ohmori, Y., and Hamilton, T. A. (1995) J. Immunol. 154, 5235–5244 appears to be critical for the cytotoxic synergism with TNFa. 8. Johnson, D. R., and Pober, J. S. (1994) Mol. Cell. Biol. 14, 1322–1332 9. Collins, T., Read, M. A., Neish, A. S., Whitley, M. Z., Thanos, D., and Maniatis, Our results indicate that IFNg/TNFa induces death signal- T. (1995) FASEB J. 9, 899 –909 ing in ME-180 cells regardless of caspase activation and acti- 10. Ruggiero, V., Tavernier, J., Fiers, W., and Baglioni, C. (1986) J. Immunol. 136, vation of caspases determines the final mode of cell death 2445–2450 11. Tsujimoto, M., Feinman, R., and Vilcek, J. (1986) J. Immunol. 137, 2272–2276 (apoptosis versus necrosis). These results suggest that IFNg/ 12. Aggarwal, B. B., and Eessalu, T. E. (1987) J. Biol. Chem. 262, 10000 –10007 TNFa-induced apoptotic and necrotic death signaling path- 13. Shimizu, S., Eguchi, Y., Kamiike, W., Itoh, Y., Hasegawa, J., Yamabe, K., Otsuki, Y., Matsuda, H., and Tsujimoto, Y. (1996) Cancer Res. 56, ways have common signaling components, and the mode of cell 2161–2166 death depends on distinct signaling events such as caspase 14. Brockman, J. A., Scherer, D. C., McKinsey, T. A., Hall, S. M., Qi, X., Lee, W. Y., activation. A similar dual pathway in cell death has been and Ballard, D. W. (1995) Mol. Cell. Biol. 15, 2809 –2818 15. Lee, K.-Y., Chang, W., Qiu, D., Kao, P. N., and Rosen, G. D. (1999) J. Biol. reported in L929 cells transfected with Fas cDNA (35). Ligation Chem. 274, 13451–13455 of Fas with anti-Fas antibody induced apoptosis of these cells. 16. Ballard, D. W., Dixon, E. P., Peffer, N. J., Bogerd, H., Doerre, S., Stein, B., and However, pretreatment with z-VAD, which inhibits activation Greene, W. C. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1875–1879 17. Ekner, R., Ewen, M. E., Newsome, D., Gerdes, M., DeCaprio, J. A., Lawrence, of caspases, resulted in necrotic death. Moreover, necrosis of J. B., and Livingston, D. M. (1994) Genes Dev. 8, 869 – 884 Fas-expressing L929 cells was inhibited by reactive oxygen 18. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W. (1989) Nucleic Acids Res. 17, 6419 intermediate (ROI) scavengers such as butylated hydroxya- 19. Wang, C. Y., Mayo, M. W., Korneluk, R. G., Goeddel, D. V., and Baldwin, A. S., nisol, indicating the involvement of ROI generation in necrotic Jr. (1998) Science 281, 1680 –1683 cell death pathway. Butylated hydroxyanisol, however, did not 20. Beg, A. A., and Baltimore, D. (1996) Science 274, 782–784 21. Wang, C. Y., Mayo, M. W., and Baldwin, A. S., Jr. (1996) Science 274, 784 –787 block IFNg/TNFa-induced ME-180 cell death in the presence of 22. Liu, Z. G., Hsu, H., Goeddel, D. V., and Karin, M. (1996) Cell 87, 565–576 caspase inhibitors (data not shown), suggesting distinct signal 23. Palombella, V. J., Rando, O. J., Goldberg, A. L., and Maniatis, T. (1994) Cell transduction between the two cell types. Nevertheless, dual 78, 773–785 24. Dong, Y., Rohn, W. M., and Benveniste, E. N. (1999) J. Immunol. 162, pathways of death signaling appear to be present in the two 4731– 4739 cells, and it will be of great interest to see if this type of 25. Stark, G. D., Kerr, I. M., Williams, B. R. G., Silverman, R. H., and Schreiber, R. D. (1998) Annu. Rev. Biochem. 67, 227–264 response could be found in other cell types exposed to similar or 26. Tanaka, N., Ishihara, M., Kitagawa, M., Harada, H., Kimura, T., Matsuyama, different death signals. Whether a cell undergoes apoptosis or T., Lamphier, M. S., Aizawa, S., Mak, T. W., and Taniguchi, T. (1994) Cell necrosis by a given stimulus may be determined by intracellu- 77, 829 – 839 27. Tamura, T., Ishihara, M., Lamphier, M. S., Tanaka, N., Oishi, I., Aizawa, S., lar milieu (36 –38). Intracellular levels of ATP were reported to Matsuyama, T., Mak, T. W., Taki, S., and Taniguchi, T. (1995) Nature 376, be a determinant of manifestation of cell death (apoptosis ver- 596 –599 28. Kano, A., Haruyama, T., Akaike, T., and Watanabe, Y. (1999) Biochem. Bio- sus necrosis) (36). Also, the fact that Bcl-2 blocks both apoptotic phys. Res. Commun. 257, 672– 677 and necrotic cell death supports the presence of common sig- 29. Neish, A. S., Read, M. A., Thanos, D., Pine, R., Maniatis, T., and Collins, T. naling components between the two death-signaling pathways (1995) Mol. Cell. Biol. 15, 2558 –2569 30. Drew, P. D., Franzoso, G., Becker, K. G., Bours, V., Carlson, L. M., Siebenlist, (37, 38). Although our work cannot provide detailed biochemi- U., and Ozato, K. (1995) J. Interferon Cytokine Res. 15, 1037–1045 cal mechanisms of cell death machinery in ME-180 cells, our 31. Saura, M., Zaragoza, C., Bao, C., McMillan, A., and Lowenstein, C. J. (1999) J. studies point out the existence of common components between Mol. Biol. 289, 459 – 471 32. Sheppard, K. A., Rose, D. W., Haque, Z. K., Kurokawa, R., McInerney, E., apoptotic and necrotic death signaling and the role of caspases Westin, S., Thanos, D., Rosenfeld, M. G., Glass, C. K., and Collins, T. (1999) in determining the type of cell death, which may help under- Mol. Cell. Biol. 19, 6367– 6378 33. Hottiger, M. O., Felzien, L. K., and Nabel, G. J. (1998) EMBO J. 17, 3124 –3134 stand the general cell death mechanism. 34. Cheshire, J. L., and Baldwin, A. S., Jr. (1997) Mol. Cell. Biol. 17, 6746 – 6754 In conclusion, we report a novel signal transduction of IFNg/ 35. Vercammen, D., Brouckaert, G., Denecker, G., Van de Craen, M., Declercq, W., TNFa synergism in the induction of ME-180 cell apoptosis: Fiers, W., and Vandenabeele, P. (1998) J. Exp. Med. 188, 919 –930 36. Eguchi, Y., Shimizu, S., and Tsujimoto, Y. (1997) Cancer Res. 57, 1835–1840 37. Kane, D. J., Sarafian, T. A., Anton, R., Hahn, H., Gralla, E. B., Valentine, J. S., K. Suk, I. Chang, Y.-H. Kim, J. Y. Kim, and M.-S. Lee, unpublished Ord, T., and Bredesen, D. E. (1993) Science 262, 1274 –1277 data. 38. Cory, S. (1995) Annu. Rev. Immunol. 13, 513–543 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Biological Chemistry Unpaywall

Interferon γ (IFNγ) and Tumor Necrosis Factor α Synergism in ME-180 Cervical Cancer Cell Apoptosis and Necrosis

Journal of Biological ChemistryApr 1, 2001

Loading next page...
 
/lp/unpaywall/interferon-ifn-and-tumor-necrosis-factor-synergism-in-me-180-cervical-qdbkVYJT36

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0021-9258
DOI
10.1074/jbc.m007646200
Publisher site
See Article on Publisher Site

Abstract

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 276, No. 16, Issue of April 20, pp. 13153–13159, 2001 © 2001 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Interferon g (IFNg) and Tumor Necrosis Factor a Synergism in ME-180 Cervical Cancer Cell Apoptosis and Necrosis IFNg INHIBITS CYTOPROTECTIVE NF-kB THROUGH STAT1/IRF-1 PATHWAYS* Received for publication, August 22, 2000, and in revised form, December 15, 2000 Published, JBC Papers in Press, January 18, 2001, DOI 10.1074/jbc.M007646200 ¶i i Kyoungho Suk‡§ , Inik Chang‡ , Yun-Hee Kim**, Sunshin Kim**, Ja Young Kim**, Hocheol Kim §, and Myung-Shik Lee‡**‡‡ From the ‡Clinical Research Center, Samsung Biomedical Research Institute and **Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul 135-710, Korea and §Graduate School of East-West Medical Science, Kyunghee University, Seoul, 130-701, Korea We investigated the molecular mechanism of the syn- wide variety of biological activities such as induction of septic ergism between interferon g (IFNg) and tumor necrosis shock, activation of local inflammatory responses, and fever factor a (TNFa) documented in a variety of biological generation as an endogenous pyrogen (1). TNFa also kills var- occasions such as tumor cell death and inflammatory ious tumor cell lines in vitro and mediates anti-tumor effect in responses. IFNg/TNFa synergistically induced apoptosis vivo (2). TNFa exerts its biological effects by binding to two of ME-180 cervical cancer cells. IFNg induced STAT1 types of cell surface receptors with molecular masses of 55 kDa phosphorylation and interferon regulatory factor 1 (p55) and 75 kDa (p75). TNFa cytotoxicity is mostly mediated (IRF-1) expression. Transfection of phosphorylation-de- by p55 receptors (3). After the ligation of p55 receptors, a fective STAT1 inhibited IFNg/TNFa-induced apoptosis, canonical apoptotic signal transduction pathway is initiated. whereas IRF-1 transfection induced susceptibility to The cytoplasmic death domain of p55 receptor interacts with TNFa. Dominant-negative IkBa transfection sensitized the death domain of intracellular adapter molecules such as ME-180 cells to TNFa. IFNg pretreatment attenuated TRADD (TNF receptor-associated death domain protein) and TNFa- or p65-induced NF-kB reporter activity, whereas FADD (Fas-associated death domain protein), which leads to it did not inhibit p65 translocation or DNA binding of the activation of initiator caspases (4). This, in turn, triggers NF-kB. IRF-1 transfection alone inhibited TNFa-in- the caspase cascade and ultimately results in apoptotic cell duced NF-kB activity, which was reversed by coactiva- death. tor p300 overexpression. Caspases were activated by In many cases, the anti-tumor effect of TNFa was enhanced IFNg/TNFa combination; however, caspase inhibition by IFNg (5) or metabolic inhibitors such as cycloheximide and did not abrogate IFNg/TNFa-induced cell death. In- actinomycin D (6). Although these metabolic inhibitors are stead, caspase inhibitors directed IFNg/TNFa-treated believed to block the synthesis of cytoprotective proteins, the ME-180 cells to undergo necrosis, as demonstrated by effects of IFNg might be mediated by the induction of new Hoechst 33258/propidium iodide staining and electron microscopy. Taken together, our results indicate that proteins that increase the sensitivity of target cells to TNFa. IFNg and TNFa synergistically act to destroy ME-180 IFNg/TNFa synergism also has been reported in biological tumor cells by either apoptosis or necrosis, depending responses other than tumor cell killing. For instance, the two on caspase activation, and STAT1/IRF-1 pathways initi- cytokines synergistically up-regulated the expression of nu- ated by IFNg play a critical role in IFNg/TNFa syner- merous genes, including ICAM-1 (intercellular adhesion mole- gism by inhibiting cytoprotective NF-kB. IFNg/TNFa cule 1), IP-10, and major histocompatibility complex class I synergism appears to activate cell death machinery in- heavy chain (7–9). However, the molecular mechanism of the dependently of caspase activation, and caspase activa- synergism between the two cytokines is not clearly understood. tion seems to merely determine the mode of cell death. It has been reported that IFNg increases the expression of TNFa receptors (10). However, because the sensitivity of the cells to TNFa is not simply correlated with the level of TNFa The pleiotropic proinflammatory cytokine TNFa exerts a receptor expression (11, 12), up-regulation of TNFa receptor alone does not adequately explain the cytokine synergism in the anti-tumor action. * This work was supported by National Research Laboratory Grants In the current work, we utilized ME-180 human cervical 2000-N-NL-01-C-232 from the Korea Institute of Science and Technol- cancer cells to investigate the molecular mechanism of syner- ogy Evaluation and Planning and by Science Research Center Grants from Korea Science and Engineering Foundation. The costs of publica- gistic anti-tumor effects of IFNg/TNFa. We also studied the tion of this article were defrayed in part by the payment of page role of caspase activation in ME-180 cell death by IFNg/TNFa charges. This article must therefore be hereby marked “advertisement” synergism. Our results indicate that 1) IRF-1 induction after in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. STAT1 activation by IFNg plays a central role in synergistic Supported by Brain Korea 21 project from the Ministry of Educa- tion, Korea. These authors contributed equally to this work. ‡‡ Recipient of Juvenile Diabetes Foundation International Research benzyloxycarbonyl-Val-Ala-Asp(OCH )-CH -fluoromethyl ketone; BD- 3 2 Grant 1-1999-760). To whom correspondence should be addressed. Tel.: fmk, t-butoxycarbonyl-Asp(OCH )-CH -fluoromethyl ketone; z-DEVD- 3 2 82-2-3410-3436; Fax: 82-2-3410-3849; E-mail: mslee@smc.samsung. fmk, benzyloxycarbonyl-Asp(OCH )-Glu(OCH )-Val-Asp(OCH )-CH - 3 3 3 2 co.kr. fluoromethyl ketone; z-IETD-fmk, benzyloxycarbonyl-Ile-Glu(OCH )- The abbreviations used are: TNF, tumor necrosis factor; IFN, int- Thr-Asp(Ome)-CH -fluoromethyl ketone; MTT, 3-[4,5-dimethylthiazol- erferon; IRF, interferon regulatory factor; STAT, signal transducer and 2-yl]-2,5-diphenyltetrazolium bromide; Ac, acetyl; AMC, amidome- activator of transcription; PI, propidium iodide; z-VAD-fmk, thylcoumarin. This paper is available on line at http://www.jbc.org 13153 This is an Open Access article under the CC BY license. 13154 IFNg/TNFa Synergism in ME-180 Cell Death anti-human phospho-STAT1, New England Biolabs) and horseradish tumor cell death by IFNg/TNFa,2)IFNg-induced IRF-1 inhib- peroxidase-conjugated secondary antibodies (anti-rabbit IgG, Amer- its cytoprotective NF-kB transactivation, 3) IFNg/TNFa in- sham Pharmacia Biotech), followed by ECL detection (Amersham Phar- duces ME-180 cell death regardless of caspase activation, and macia Biotech). caspase activation dictates only the mode of cell death between Transient Transfection—ME-180 cells in 6-well plates were co-trans- apoptosis and necrosis. fected with 1 mg of human STAT1 cDNA, dominant-negative mutant STAT1 cDNA (kindly provided by Dr. Hirano, Osaka University, Ja- EXPERIMENTAL PROCEDURES pan), human IRF-1 cDNA (kindly provided by Dr. Taniguchi, Univer- Cell Line and Reagents—ME-180 cervical cancer cell line was ob- sity of Tokyo), or phosphorylation-defective dominant-negative mutant tained from ATCC (Manassas, VA) and grown in Dulbecco’s modified IkBa (14) together with 0.2 mgof lacZ gene (pCH110, Amersham Phar- Eagle’s medium containing 10% fetal bovine serum, 2 mM glutamine, macia Biotech) using LipofectAMINE reagent (Life Technologies, Inc.). and penicillin-streptomycin (Life Technologies, Inc.). Recombinant hu- 48 h after the transfection, cells were treated with cytokines. After man IFNg was purchased from R&D Systems (Minneapolis, MN). Re- another 48 h, the cells were fixed with 0.5% glutaraldehyde for 10 min combinant human TNFa was generously provided by Dr. T. H. Lee at room temperature and stained with X-gal (5-bromo-4-chloro-3-indo- (Yonsei University, Seoul, Korea). Caspase inhibitors (z-VAD-fmk, ben- lyl b-D-galactopyranoside; 1 mg/ml) in 4 mM potassium ferricyanide, 4 zyloxycarbonyl-Val-Ala-Asp(OCH )-CH -fluoromethyl ketone; BD-fmk, mM potassium ferrocyanide, 2 mM magnesium chloride at 37 °C for 3 2 t-butoxycarbonyl-Asp(OCH )-CH F; z-DEVD-fmk, benzyloxycarbonyl- detection of blue cells. At least 200 blue cells were counted for each 3 2 Asp(OCH )-Glu(OCH )-Val-Asp(OCH )-CH -fluoromethyl ketone; z-IETD- experiment, and transfection efficiency was 10 –35%. Results were pre- 3 3 3 2 fmk, benzyloxycarbonyl-Ile-Glu(OCH )-Thr-Asp(OCH )-CH -fluoromethyl sented as means 6 S.E. (n 5 3). 3 3 2 ketone) were purchased from Enzyme Systems (Livermore CA), and cathep- NF-kB Reporter Assays—NF-kB reporter activity was measured us- sin B inhibitor FA (benzyloxycarbonyl-Phe-Ala-CH -fluoromethyl ketone) ing the dual-luciferase reporter assay system (Promega, Madison, WI). and MG-132 (carbobenzoxyl-leucinyl-leucinyl-leucinal-H, also called Z-LLL) In brief, ME-180 cells in 12-well plates were co-transfected with 0.5 mg were from Calbiochem. All other chemicals were obtained from Sigma, of NF-kB-responsive reporter gene construct carrying two copies of kB unless stated otherwise. sequences linked to luciferase gene (IgGk NF-kB-luciferase, generously Assessment of Cytotoxicity by 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphe- provided by Dr. G. D. Rosen, Stanford University, Stanford, CA) (15) nyltetrazolium Bromide (MTT) Assay—Cells (3 3 10 /well) were seeded together with 0.1 mgof Renilla luciferase gene under hamster sarcoma in 96-well plates and treated with various combinations of cytokines for virus thymidine kinase promoter (pRL-TK, Promega) using Lipo- the indicated time periods. The optimal concentrations of the cytokines fectAMINE reagent (Life Technologies, Inc.). 24 h after the transfection, for the cytotoxic action were 100 units/ml for IFNg and 10 ng/ml for cells were treated with cytokines. After 5 h, activities of firefly lucifer- TNFa. In some experiments, cells were pretreated with caspase inhib- ase and Renilla luciferase in transfected cells were measured sequen- itors or MG-132 for 1 h before cytokine treatment. After cytokine tially from a single sample using the dual-luciferase reporter assay treatment, the medium was removed, and MTT (0.5 mg/ml) was added, system (Promega). Results were presented as firefly luciferase activity followed by incubation at 37 °C for2hinCO incubator. After a brief normalized to Renilla luciferase activity. In some experiments, cells centrifugation, supernatants were carefully removed, and Me SO was were co-transfected before cytokine treatment with NF-kB p65 (16) or added. After insoluble crystals were completely dissolved, absorbance coactivator p300 expression plasmid (0.5 mg; kindly provided by Dr. at 540 nm was measured using a Thermomax microplate reader (Mo- Livingston, Harvard Medical School, Boston, MA) (17) along with NF- lecular Devices). Results were presented as means 6 S.E. (n 5 3). kB-responsive reporter plasmid (0.5 mg) and pRL-TK (0.1 mg). Results Morphological Analysis of Apoptotic Cells—Morphological changes in were presented as means 6 S.E. (n 5 3). the nuclear chromatin of cells undergoing apoptosis were detected by Immunofluorescence Staining—ME-180 cells seeded onto chamber staining with 2.5 mg/ml bisbenzimide Hoechst 33258 fluorochrome (Cal- slides (Lab-Tek, Nalge Nunc International, Naperville, IL) were fixed in biochem), followed by examination on a fluorescence microscope. In 4% paraformaldehyde for 30 min at room temperature and then in cold some experiments, cytokine-treated cells were double-stained with pro- methanol for 10 min at 220 °C. Fixed cells were permeabilized in 0.1% pidium iodide (PI, 2.5 mg/ml) and Hoechst 33258 (2.5 mg/ml) to distin- Triton X-100, 0.1% sodium citrate for 3 min at 4 °C and then sequen- guish apoptotic cells from necrotic cells. Intact blue nuclei, condensed/ tially incubated with mouse anti-p65 antibody (Santa Cruz Biotechnol- fragmented blue nuclei, condensed/fragmented pink nuclei, and intact ogy, Santa Cruz, CA), biotinylated anti-mouse IgG, and streptavidin- pink nuclei were considered viable, early apoptotic, late apoptotic, and fluorescein isothiocyanate. Stained cells were examined on a necrotic cells, respectively (13). Transmission electron microscopy was fluorescent microscope. carried out essentially as previously described (13). In brief, cells were Electrophoretic Mobility Shift Assay—Nuclear extracts were pre- fixed in 4% glutaraldehyde, 1% paraformaldehyde, 0.2 M phosphate, pH pared from ME-180 cells treated with cytokines as previously described 7.2, at 4 °C for 2 h. After two washes in 0.2 M phosphate, the cell pellet (18). Synthetic double-strand oligonucleotides of consensus NF-kB was post-fixed with 2% OsO in the same buffer for 30 min. The pellet binding sequence, GAT CCC AAC GGC AGG GGA (Promega), were was dehydrated in ethanol and then in 100% propylene oxide, followed end-labeled with [g- P]ATP using T4 polynucleotide kinase. Nuclear by embedding overnight at 37 °C for another 3 days at 60 °C. Ultrafine extract was incubated with the labeled probe in the presence of poly- sections were cut and examined on an electron microscope (Hitachi (dI-dC) in a binding buffer containing 20 mM HEPES at room temper- H7100, 75 kV). ature for 30 min. For supershift assays, a total of 0.2 mg of antibodies DNA Ploidy Analysis—Cells were suspended in phosphate-buffered against p65 or p50 subunit of NF-kB were included in the reaction. saline, 5 mM EDTA and fixed by adding 100% ethanol dropwise. RNase DNA-protein complexes were resolved by electrophoresis in a 5% nonde- A (40 mg/ml) was added to resuspended cells, and the incubation was naturing polyacrylamide gel, dried, and visualized by autoradiography. carried out at room temperature for 30 min. PI (50 mg/ml) was then added for flow cytometric analyses. RESULTS Assessment of Caspase Activity—Caspase-3- or -8-like activity was IFNg and TNFa Synergistically Induced the Apoptosis of measured using a caspase assay kit (Pharmingen, San Diego, CA) ME-180 Cells—First we screened several tumor cell lines to according to the supplier’s instruction. In brief, caspase-3 or -8 fluoro- genic substrates (Ac-DEVD-AMC or Ac-IETD-AMC) were incubated assess their sensitivity to IFNg/TNFa-induced cytotoxicity with cytokine-treated cell lysates for1hat37 °C, then AMC liberated (data not shown). Cytotoxic synergism between IFNg and from Ac-DEVD-AMC or Ac-IETD-AMC was measured using a fluoro- TNFa was most evident in ME-180 cells. Although either cy- metric plate reader with an excitation wavelength of 380 nm and an tokine alone exhibited no significant cytotoxicity, the combina- emission wavelength of 420- 460 nm. tion of the two cytokines significantly reduced ME-180 cell Western Blot Analysis—Cells were lysed in triple-detergent lysis viability (Fig. 1A). The cytokine cytotoxicity was dependent on buffer (50 mM Tris-Cl, pH 8.0, 150 mM NaCl, 0.02% sodium azide, 0.1% SDS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 1 mM phenylmeth- the dose of IFNg used. However, concentration higher than 100 ylsulfonyl fluoride). Protein concentration in cell lysates was deter- units/ml did not further increase the cytotoxicity (Table I). The mined using the Bio-Rad protein assay kit. An equal amount of protein reduction of cell viability was due to apoptosis as demonstrated for each sample was separated by 10 or 12% SDS-polyacrylamide gel by Hoechst 33258 staining and DNA ploidy analysis. IFNg/ electrophoresis and transferred to Hybond ECL nitrocellulose mem- TNFa treatment induced nuclear condensation and fragmen- branes (Amersham Pharmacia Biotech). The membranes were blocked tation (Fig. 1B) and led to the appearance of sub-diploid cells with 5% skim milk and sequentially incubated with primary antibodies (rabbit anti-human IRF-1, Santa Cruz; rabbit anti-human STAT1 and (Fig. 1C), which are hallmarks of apoptotic cells. DNA ploidy IFNg/TNFa Synergism in ME-180 Cell Death 13155 TABLE II Cytotoxic effects of sequential treatment of cytokines a b Treatment % Viability None 100 IFNg 1 TNFa (48 h) 22.1 6 3.1 IFNg (48 h) 98.2 6 3.4 IFNg (24 h) and then TNFa (48 h) 52.3 6 2.4 TNFa (48 h) 101.5 6 2.9 TNFa (24 h) and then IFNg (48 h) 90.3 6 3.3 ME-180 cells were treated with cytokines as indicated, either simul- taneously or sequentially. Treatment with IFNg (100 units/ml) for 24 h followed by TNFa (10 ng/ml) treatment for 48 h induced a significant cytotoxicity. However, sequential treatment with the two cytokines in a reverse order did not significantly affect ME-180 cell viability, indicat- ing the priming role of IFNg in TNFa-induced ME-180 cell death. Cell viability was assessed by MTT assays. Viability of untreated cells was set to 100%. Results are means 6 S.E. (n 5 3). FIG.2. IFNg activates STAT1 (A) and induces IRF-1 expression (B) in ME-180 cells. Western blot analyses demonstrated that treat- FIG.1. IFNg/TNFa synergistically induces ME-180 cell apopto- ment of ME-180 cells with IFNg induced STAT1 expression (24-h treat- sis. A combination of IFNg (100 units/ml) and TNFa (10 ng/ml), but not ment) as well as its phosphorylation (30-min treatment) (A). IFNg also either cytokine alone, induced ME-180 cell death. Cell viability was induced IRF-1 expression at 24 h after treatment, and the expression assessed by MTT assays after treatment with the cytokines for 48 h (A). was further increased at 48 h after the treatment (B). However, TNFa Induction of ME-180 cell death was due to apoptosis, as demonstrated alone did not change the expression of either STAT1 or IRF-1. C, by chromatin condensation in Hoechst 33258 staining (B) or the ap- untreated control; I, IFNg (100 units/ml); T, TNFa (10 ng/ml); D, IFNg pearance of sub-G peak in flow cytometric analyses (C) at 24 h after plus TNFa. cytokine treatment. with IFNg did not have the same effects, indicating that IFNg TABLE I confers susceptibility to TNFa on ME-180 cells through induc- Dose response of cytokine cytotoxicity tion or up-regulation of certain genes in ME-180 cells. Because Treatments % Viability STAT1 and IRF-1 are known to be canonical intracellular sig- None 100 nal-transducing molecules in IFNg signaling, we investigated IFNg (1 unit/ml) 1 TNFa 95.6 6 3.6 the involvement of STAT1/IRF-1-signaling pathways in IFNg/ IFNg (10 units/ml) 1 TNFa 76.2 6 2.5 TNFa synergism on ME-180 cell apoptosis. IFNg, but not IFNg (100 units/ml) 1 TNFa 23.4 6 3.1 TNFa, induced phosphorylation of STAT1 and up-regulated IFNg (1000 units/ml) 1 TNFa 24.7 6 2.8 IRF-1 expression in ME-180 cells (Fig. 2). Furthermore, the ME-180 cells were treated with increasing concentrations of IFNg transfection of phosphorylation-defective dominant-negative and TNFa (10 ng/ml) for 48 h, and then cell viability was assessed by MTT assays. Viability of untreated cells was set to 100%. Results are mutant of STAT1 significantly inhibited IFNg/TNFa-induced the means 6 S.E. (n 5 3). IFNg alone at all concentrations tested did ME-180 cell death, indicating that IFNg-induced STAT1 acti- not have a significant cytotoxicity (data not shown). One hundred vation is critical for the induction of TNFa susceptibility (Fig. units/ml IFNg and 10 ng/ml TNFa were the optimal concentrations for 3A). We next asked whether IRF-1, a downstream mediator of cytotoxic assays. STAT1, is responsible for the priming effects of IFNg. Trans- fection of IRF-1 conferred TNFa susceptibility on ME-180 cells in a dose-dependent manner, indicating a central role for IRF-1 assays also indicated that the effect of IFNg/TNFa was not due in the sensitization of ME-180 cells to TNFa-induced apoptosis to the growth arrest as was shown by the absence of decrease in (Fig. 3, B and C). the percentage of cells in the S phase. Inhibition of Cytoprotective NF-kB Activity by IFNg—TNFa is IFNg/TNFa Synergism Involved IFNg-induced STAT1 Acti- known to initiate both death and survival signals, and recent stud- vation and IRF-1 Induction—Based on our results that the ies on TNFa-induced survival signal suggested an important role of combination of IFNg and TNFa, but not either cytokine alone, NF-kB activation (19 –22). Thus, we investigated how IFNg in- induced ME-180 cell death, we explored the possibility that duces susceptibility to TNFa-induced cytotoxicity by examining the IFNg sensitizes ME-180 cells to TNFa-mediated cytotoxicity. role of NF-kB in ME-180 cell death and its possible regulation by This was first tested by sequential treatment of ME-180 cells IFNg. Treatment of ME-180 cells with a proteasome inhibitor (MG- with the two cytokines. After IFNg treatment, TNFa alone was 132), which is known to inhibit NF-kB activation (23), rendered the sufficient to induce a significant cytotoxicity in ME-180 cells cells sensitive to TNFa-induced apoptosis (Fig. 4A), suggesting the (Table II). However, sequential treatment with TNFa and then cytoprotective role of NF-kB. Also, upon the transfection of phos- 13156 IFNg/TNFa Synergism in ME-180 Cell Death FIG.3. A key role for STAT1/IRF-1 signaling in IFNg/TNFa synergism. A, transient transfection of phosphorylation- defective STAT1 dominant-negative mu- tant (DN STAT1) significantly inhibited IFNg/TNFa cytotoxicity, as demonstrated by counting blue cells co-expressing lacZ at 48 h after cytokine treatment (IFNg, 100 units/ml; TNFa, 10 ng/ml). B, trans- fection of IRF-1 cDNA (1 mg) induced sus- ceptibility to TNFa. In contrast to empty vector (pcDNA3) transfectants, treatment of IRF-1 transfectants with TNFa alone for 48 h significantly decreased the num- ber of blue cells. C, the effects of IRF-1 were dependent upon the dose of IRF-1 cDNA (0.1, 0.5, 1, and 2 mg) used in the transient transfection. The number of blue cells upon transfection with an empty vector without TNFa treatment was set to 100%. FIG.4. Inhibition of cytoprotective NF-kBbyIFNg. A, inhibition of NF-kB by proteasome inhibitor MG-132 sensi- tized ME-180 cells to TNFa. ME-180 cells were treated with either MG-132 (0.5 mM) alone or in combination with TNFa (10 ng/ml) for 48 h, and then cell viability was assessed by MTT assays. B, inhibition of NF-kB by transfection of dominant-nega- tive mutant IkBa (DN IkBa) also ren- dered ME-180 cells sensitive to TNFa treatment. Viability of ME-180 cells co- transfected with dominant-negative IkBa and lacZ was significantly decreased by TNFa treatment (24 h), in contrast to the cells co-transfected with an empty vector (pcDNA3) and lacZ. C and D, NF-kB re- porter assays revealed that pretreatment (24 h, 100 units/ml) of ME-180 cells with IFNg inhibited TNFa-induced NF-kB ac- tivity (C). IFNg treatment (48 h) also in- hibited NF-kB reporter activity induced by p65 transfection (NF-kB p65)(D). Transiently transfected cells were treated with cytokines for the indicated time period before NF-kB reporter assays (C and D). phorylation-defective dominant-negative mutant IkBa, TNFa of IFNg on NF-kB. Transfection of IRF-1 alone was sufficient to alone induced a significant cytotoxicity, further supporting the cy- inhibit TNFa-induced NF-kB activity, indicating a central role of toprotective role of NF-kB (Fig. 4B). NF-kB reporter assays indi- IRF-1 in the inhibition of NK-kB transactivation by IFNg (Fig. 7A). cated that IFNg pretreatment attenuated TNFa-induced NF-kB We also investigated the possible mechanism of interference be- activity, suggesting that IFNg synergizes with TNFa for ME-180 tween IRF-1 and NF-kB. Transfection of p300 coactivator abro- cell apoptosis by inhibiting TNFa-induced cytoprotective NF-kB gated the inhibitory effect of IFNg treatment (Fig. 7B) or IRF-1 activity (Fig. 4C). IFNg pretreatment, however, did not inhibit transfection (Fig. 7C)onTNFa-induced NF-kB activity, suggesting nuclear translocation of p65 (Fig. 5) or DNA binding of NF-kB the possibility of coactivator competition between IFNg-induced induced by TNFa treatment (Fig. 6). Also, IFNg did not inhibit IRF-1 and TNFa-induced NF-kB. TNFa-induced degradation of IkBa (data not shown). However, Inhibition of Caspases Directed ME-180 Cells to Undergo IFNg treatment did inhibit the NF-kB reporter activity induced by Necrotic Cell Death—We next investigated whether the activa- transfection of p65 subunit of NF-kB (Fig. 4D), indicating that tion of caspases is involved in the IFNg/TNFa-induced apopto- IFNg directly inhibited NF-kB-mediated transactivation within the sis of ME-180 cells. Cytokine-induced apoptosis of ME-180 cells nuclei without affecting the nuclear translocation or DNA binding was accompanied by the activation of caspase-3-like activity, as of NF-kB. We next studied if IRF-1 mediates this inhibitory action demonstrated by the cleavage of Ac-DEVD-AMC in IFNg/ IFNg/TNFa Synergism in ME-180 Cell Death 13157 TNFa-treated cells (Fig. 8). Cytokine treatment also induced ergism in ME-180 cell apoptosis. However, dominant-negative the cleavage of Ac-IETD-AMC, indicating concurrent activation STAT1 did not completely abolish cytotoxicity by IFNg/TNFa, of caspase-8-like activity (data not shown). However, pretreat- and IRF-1 transfection could not be completely substituted for ment with broad-spectrum caspase inhibitors such as z-VAD- IFNg. IFNg induces STAT1 as well as IRF-1, and some cellular fmk or BD-fmk failed to inhibit ME-180 cell death by IFNg/ responses to IFNg are reported to be mediated by both STAT1 TNFa synergism despite the activation of multiple caspases and IRF-1 (24 –25). Neither STAT1 or IRF-1 alone may not (Fig. 9A). Instead, IFNg/TNFa in the presence of caspase in- explain all of the priming effect of IFNg in TNFa-induced hibitors unexpectedly induced the necrosis of ME-180 cells, as death. The role of IRF-1 in the induction of apoptosis by DNA judged by the swelling of dying cells on a light microscope (data damage or IFNg has been previously suggested (26 –28), which not shown). Hoechst 33258/PI staining and electron microscopy supports the proapoptotic action of IRF-1. Previous work in our confirmed the necrosis of the cells (Fig. 9, B and C). To study laboratory also showed that IRF-1 plays a central role in IFNg/ the involvement of individual caspases in the switching process TNFa-induced apoptosis of pancreatic islet b-cells in autoim- from apoptosis to necrosis, cells were pretreated with inhibitors mune diabetes. Caspase induction has been suggested as a specific for individual caspases instead of z-VAD-fmk. Because possible downstream event after IRF-1 induction in IFNg-in- we observed the activation of caspase-3 and -8 in the cytokine- duced apoptosis (28). Although RNase protection assays re- treated ME-180 cells, we tested the effects of z-DEVD-fmk and z-IETD-fmk alone or in combination. The z-DEVD-fmk and z-IETD-fmk acted additively in conversion from apoptosis to K. Suk, I. Chang, Y.-H. Kim, S. Kim, J. Y. Kim, and M.-S. Lee, submitted for publication. necrosis, suggesting the involvement of multiple caspases in determining the mode of ME-180 cell death (Table III). DISCUSSION Here we present evidence that STAT1/IRF-1 pathways ini- tiated by IFNg play a central role in IFNg/TNFa synergism in the induction of ME-180 cell apoptosis. Transfection of domi- nant-negative STAT1 abolished IFNg/TNFa synergism, whereas transfection of IRF-1 sensitized ME-180 cells to TNFa-induced apoptosis. Thus, STAT1 activation and IRF-1 induction by IFNg appear to be important in IFNg/TNFa syn- FIG.6. No significant effects of IFNg on DNA binding of NF-kB protein. A, IFNg pretreatment (100 units/ml, 24 h) did not signifi- cantly affect TNFa-induced kB sequence binding of NF-kB proteins (lanes 4 and 5). The identity of DNA-complexed proteins was confirmed by supershift assays using antibodies (Ab) against p65 (lane 6), p50 (lane 7), or both (lane 8). B, ME-180 cells were similarly treated with increasing doses of IFNg and TNFa as indicated, and then NF-kB was detected by electrophoretic mobility shift assay. IFNg at all concentra- FIG.5. No effects of IFNg on nuclear translocation of p65 sub- unit of NF-kB. As compared with untreated control (A), TNFa treat- tions tested did not significantly influence TNFa-induced DNA binding of NF-kB, indicating that the inability of IFNg to inhibit TNFa-induced ment (45 min, 10 ng/ml) induced nuclear translocation of p65 (B), which was not affected by IFNg pretreatment (24 h, 100 units/ml) (C). DNA binding of NF-kB was not due to the low dose of IFNg used. FIG.7. IRF-1 mediates NF-kB-inhibiting effects of IFNg probably through coactivator competition. A, transfection of IRF-1 inhibited TNFa-induced NF-kB reporter activity in a manner similar to IFNg pretreatment. B, transfection of coactivator p300 abrogated IFNg-mediated inhibition of NF-kB reporter activity. Transiently transfected cells were treated with cytokines for an indicated time period before NF-kB reporter assays (IFNg, 100 units/ml; TNFa, 10 ng/ml). C, ME-180 cells were co-transfected with NF-kB reporter construct and the indicated plasmids, and then the luciferase activity was measured after 24 h. Co-transfection of coactivator p300 also abolished the IRF-1 transfection-mediated inhibition of NF-kB reporter activity. CBP, cAMP-response element-binding protein (CREB)-binding protein. 13158 IFNg/TNFa Synergism in ME-180 Cell Death FIG.8. Activation of caspase-3-like activity by IFNg/TNFa in ME-180 cells. IFNg/TNFa treatment (IFNg, 100 units/ml; TNFa,10 ng/ml) induced cleavage of DEVD-AMC, indicating activation of caspase-3-like activity. Pretreatment of ME-180 cells with z-VAD-fmk before cytokine treatment completely inhibited the caspase activity (IFNg, 100 units/ml; TNFa, 10 ng/ml). vealed that the expression of caspase-1 and -4 was up-regu- lated by IFNg treatment in ME-180 cells (data not shown), there remains yet to be determined how the increases in the expression of these caspases mediate IRF-1 action. In IFNg/ TNFa-induced death of ME-180 cells, caspases seem to be involved in determining the mode of cell death rather than decision between death and survival (see below). Although further works are necessary to completely delineate the downstream signaling pathways after STAT1/IRF-1 in IFNg/ TNFa cytotoxic synergism, our current work indicates that NF-kB is one of the targets of STAT1/IRF-1 action. We demonstrated that IFNg attenuated TNFa-induced NF-kB reporter activity in ME- 180 cells. Also, the inhibition of NF-kB either by transfection of dominant-negative IkB “super repressor” or by treatment with a proteasome inhibitor (MG-132) rendered ME-180 cells sensitive to TNFa-induced apoptosis. These results indicate that IFNg sensi- tizes ME-180 cells to TNFa-induced apoptosis by inhibiting NF-kB- mediated activation of survival signals. Furthermore, this action of IFNg was mediated by IRF-1. It has been previously reported that FIG.9. Induction of necrotic death by IFNg/TNFa in the pres- IRF-1 and NF-kB interact in vitro as well as in vivo for the coop- ence of caspase inhibitors. A, pretreatment of ME-180 cells with erative induction of inflammatory genes (29 –31). In ME-180 cells, broad spectrum caspase inhibitors such as z-VAD-fmk or BD-fmk did however, IRF-1 negatively influenced NF-kB activity. IRF-1 does not block the cytokine-induced cytotoxicity as measured by MTT assays at 48 h after the treatment (IFNg, 100 units/ml; TNFa, 10 ng/ml). B and not seem to directly interact with NF-kB because NF-kB transcrip- C, pretreatment with z-VAD-fmk switched the mode of cell death from tional activity was assessed using a reporter construct containing a apoptosis to necrosis as judged by Hoechst 33258/PI double-staining (B) kB element but not an IRF-1 response element. Thus, in ME-180 and electron microscopy (C). In Hoechst 33258/PI double-staining, cells cells, it is likely that IRF-1 indirectly affects NF-kB transcriptional with blue intact nuclei were viable cells, whereas those with blue fragmented nuclei were early apoptotic cells. Cells with pink intact activity through the regulation of other factors modulating the nuclei were necrotic cells, whereas cells with pink fragmented nuclei transcriptional activity. We also demonstrated that IFNg did not were late apoptotic cells. The values in the parentheses below the block the TNFa-induced translocation of p65 from cytosol to nu- photographs represent the percentage of apoptotic (early or late) or cleus or DNA binding of NF-kB but yet inhibited NF-kB reporter necrotic cells out of the total 500 cells counted (B). activity. These results suggest that IFNg-induced IRF-1 inhibits the nuclear events of NF-kB transactivation but not cytosolic ing the expression of multiple genes involved in the inflammatory events. Our work also showed that transfection of transcriptional responses (34). In sharp contrast, however, our work disclosed that coactivator p300 abolished the inhibition of NF-kB reporter activity IFNg inhibited TNFa-induced NF-kB in ME-180 cells. This novel by IFNg. Transcriptional activation by NF-kB requires multiple signaling pathway of synergism between IFNg/TNFa involving coactivators (32). It has been recently reported that the intracellu- competition between IRF-1 and NF-kB for p300 coactivator may lar amount of the coactivator p300 is limited compared with other not be generalized to other cell types, considering previous reports transcriptional factors and that competition for p300 may regulate showing different signaling patterns in response to IFNg/TNFa transcriptional activity (33). Thus, it is possible that IFNg-induced (34). The same stimulus seems to activate distinct signaling path- IRF-1 competes with TNFa-induced NF-kB for the common coac- ways depending on the cell types. Because of this discrepancy in tivator(s) such as p300, and this competition may be responsible for signal transduction pathways, the final outcome would be different the inhibition of NF-kB transactivation. Then what are the target among different cell types. Some cells would undergo death by genes that are induced by NF-kB and are subject to the inhibitory IFNg/TNFa, whereas other cells may be activated by IFNg/TNFa action of IRF-1? Recently, a role of TNF receptor-associated factor to participate in inflammatory responses. 1 (TRAF2), TRAF2, c-IAP1 (inhibitor of apoptosis (IAP)) and cIAP2 Because IFNa is also known to activate the STAT1-signaling was reported in anti-apoptosis mediated by NF-kB (19). These are pathway, we investigated whether IFNa also synergizes with possible candidates for such target genes. Another puzzling point is TNFa to destroy ME-180 cells. Our results indicated that IFNa what determines how IFNg acts on NF-kB. Previously, IFNg has and TNFa synergistically induced ME-180 cell death, and this been shown to increase TNFa-induced NF-kB activation in enhanc- was accompanied by the activation of STAT1 and inhibition of IFNg/TNFa Synergism in ME-180 Cell Death 13159 TABLE III IFNg synergized with TNFa for apoptosis induction by activat- Effects of various caspase inhibitors on the conversion of ME-180 cell ing STAT1/IRF-1 pathway. We also present evidence that death from apoptosis to necrosis NF-kB activation is a survival signal in TNFa-treated ME-180 a % Viable % Necrotic % Apoptotic Treatment cells, and IFNg inhibits this survival mechanism, resulting in b b b cells cells cells synergistic cytotoxicity with TNFa. Moreover, the mode of ME- None 94.7 3.1 2.2 180 cell death by IFNg/TNFa synergism was dictated by IFNg 1 TNFa 14.7 8.6 76.7 caspase activation. The novel mechanism of IFNg/TNFa syn- IFNg 1 TNFa 1 z-VAD 16.5 78.5 5.0 ergism presented here may also be applicable to other circum- IFNg 1 TNFa 1 DEVD 16.8 25.2 58.0 IFNg 1 TNFa 1 IETD 20.4 34.5 45.1 stances, where a similar cytokine synergism could be found IFNg 1 TNFa 1 DEVD 1 IETD 27.6 55.2 17.2 such as autoimmune destruction of self tissues by cytokines. IFNg 1 TNFa 1 FA 11.2 10.5 78.3 Acknowledgments—We thank Drs. Kye Young Lee, Tae H. Lee, Jae ME-180 cells were pretreated with caspase inhibitors indicated (100 W. Lee, Minho Shong, Soo Young Lee, Il-Seon Park, and Young S. Ahn mM) for 1 h before cytokine treatment for 48 h (IFNg, 100 units/ml; for insightful discussions and technical help. TNFa, 10 ng/ml). Cathepsin B inhibitor, FA, benzyloxycarbonyl-Phe- Ala-CH -fluoromethyl ketone (FA) was used as a negative control. Percentage of viable, necrotic, or apoptotic cells was assessed by REFERENCES double-staining with Hoechst 33258 and PI. 1. Fiers, W. (1991) FEBS Lett. 285, 199 –212 2. Beyaert, R., and Fiers, W. (1994) FEBS Lett. 340, 9 –16 NF-kB reporter activity by IFNa in a manner similar to IFNg. 3. Schulze-Osthoff, K., Ferrari, D., Los, M., Wesselborg, S., and Peter, M. E. (1998) Eur. J. Biochem. 254, 439 – 459 Thus, the cytotoxic priming role of IFNg in IFNg/TNFa syner- 4. Hsu, H., Shu, H. B., Pan, M. G., and Goeddel, D. V. (1996) Cell 84, 299 –308 gism presented in the current study does not seem to be re- 5. Sugarman, B. J., Aggarwal, B. B., Hass, P. E., Figari, I. S., Palladino, M. A., Jr., and Shepard, H. M. (1985) Science 230, 943–945 stricted to IFNg. Rather, the STAT1/IRF-1-signaling pathway 6. Kirstein, M., Fiers, W., and Baglioni, C. (1986) J. Immunol. 137, 2277–2280 that can be initiated by either type I or type II interferon 7. Ohmori, Y., and Hamilton, T. A. (1995) J. Immunol. 154, 5235–5244 appears to be critical for the cytotoxic synergism with TNFa. 8. Johnson, D. R., and Pober, J. S. (1994) Mol. Cell. Biol. 14, 1322–1332 9. Collins, T., Read, M. A., Neish, A. S., Whitley, M. Z., Thanos, D., and Maniatis, Our results indicate that IFNg/TNFa induces death signal- T. (1995) FASEB J. 9, 899 –909 ing in ME-180 cells regardless of caspase activation and acti- 10. Ruggiero, V., Tavernier, J., Fiers, W., and Baglioni, C. (1986) J. Immunol. 136, vation of caspases determines the final mode of cell death 2445–2450 11. Tsujimoto, M., Feinman, R., and Vilcek, J. (1986) J. Immunol. 137, 2272–2276 (apoptosis versus necrosis). These results suggest that IFNg/ 12. Aggarwal, B. B., and Eessalu, T. E. (1987) J. Biol. Chem. 262, 10000 –10007 TNFa-induced apoptotic and necrotic death signaling path- 13. Shimizu, S., Eguchi, Y., Kamiike, W., Itoh, Y., Hasegawa, J., Yamabe, K., Otsuki, Y., Matsuda, H., and Tsujimoto, Y. (1996) Cancer Res. 56, ways have common signaling components, and the mode of cell 2161–2166 death depends on distinct signaling events such as caspase 14. Brockman, J. A., Scherer, D. C., McKinsey, T. A., Hall, S. M., Qi, X., Lee, W. Y., activation. A similar dual pathway in cell death has been and Ballard, D. W. (1995) Mol. Cell. Biol. 15, 2809 –2818 15. Lee, K.-Y., Chang, W., Qiu, D., Kao, P. N., and Rosen, G. D. (1999) J. Biol. reported in L929 cells transfected with Fas cDNA (35). Ligation Chem. 274, 13451–13455 of Fas with anti-Fas antibody induced apoptosis of these cells. 16. Ballard, D. W., Dixon, E. P., Peffer, N. J., Bogerd, H., Doerre, S., Stein, B., and However, pretreatment with z-VAD, which inhibits activation Greene, W. C. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1875–1879 17. Ekner, R., Ewen, M. E., Newsome, D., Gerdes, M., DeCaprio, J. A., Lawrence, of caspases, resulted in necrotic death. Moreover, necrosis of J. B., and Livingston, D. M. (1994) Genes Dev. 8, 869 – 884 Fas-expressing L929 cells was inhibited by reactive oxygen 18. Schreiber, E., Matthias, P., Muller, M. M., and Schaffner, W. (1989) Nucleic Acids Res. 17, 6419 intermediate (ROI) scavengers such as butylated hydroxya- 19. Wang, C. Y., Mayo, M. W., Korneluk, R. G., Goeddel, D. V., and Baldwin, A. S., nisol, indicating the involvement of ROI generation in necrotic Jr. (1998) Science 281, 1680 –1683 cell death pathway. Butylated hydroxyanisol, however, did not 20. Beg, A. A., and Baltimore, D. (1996) Science 274, 782–784 21. Wang, C. Y., Mayo, M. W., and Baldwin, A. S., Jr. (1996) Science 274, 784 –787 block IFNg/TNFa-induced ME-180 cell death in the presence of 22. Liu, Z. G., Hsu, H., Goeddel, D. V., and Karin, M. (1996) Cell 87, 565–576 caspase inhibitors (data not shown), suggesting distinct signal 23. Palombella, V. J., Rando, O. J., Goldberg, A. L., and Maniatis, T. (1994) Cell transduction between the two cell types. Nevertheless, dual 78, 773–785 24. Dong, Y., Rohn, W. M., and Benveniste, E. N. (1999) J. Immunol. 162, pathways of death signaling appear to be present in the two 4731– 4739 cells, and it will be of great interest to see if this type of 25. Stark, G. D., Kerr, I. M., Williams, B. R. G., Silverman, R. H., and Schreiber, R. D. (1998) Annu. Rev. Biochem. 67, 227–264 response could be found in other cell types exposed to similar or 26. Tanaka, N., Ishihara, M., Kitagawa, M., Harada, H., Kimura, T., Matsuyama, different death signals. Whether a cell undergoes apoptosis or T., Lamphier, M. S., Aizawa, S., Mak, T. W., and Taniguchi, T. (1994) Cell necrosis by a given stimulus may be determined by intracellu- 77, 829 – 839 27. Tamura, T., Ishihara, M., Lamphier, M. S., Tanaka, N., Oishi, I., Aizawa, S., lar milieu (36 –38). Intracellular levels of ATP were reported to Matsuyama, T., Mak, T. W., Taki, S., and Taniguchi, T. (1995) Nature 376, be a determinant of manifestation of cell death (apoptosis ver- 596 –599 28. Kano, A., Haruyama, T., Akaike, T., and Watanabe, Y. (1999) Biochem. Bio- sus necrosis) (36). Also, the fact that Bcl-2 blocks both apoptotic phys. Res. Commun. 257, 672– 677 and necrotic cell death supports the presence of common sig- 29. Neish, A. S., Read, M. A., Thanos, D., Pine, R., Maniatis, T., and Collins, T. naling components between the two death-signaling pathways (1995) Mol. Cell. Biol. 15, 2558 –2569 30. Drew, P. D., Franzoso, G., Becker, K. G., Bours, V., Carlson, L. M., Siebenlist, (37, 38). Although our work cannot provide detailed biochemi- U., and Ozato, K. (1995) J. Interferon Cytokine Res. 15, 1037–1045 cal mechanisms of cell death machinery in ME-180 cells, our 31. Saura, M., Zaragoza, C., Bao, C., McMillan, A., and Lowenstein, C. J. (1999) J. studies point out the existence of common components between Mol. Biol. 289, 459 – 471 32. Sheppard, K. A., Rose, D. W., Haque, Z. K., Kurokawa, R., McInerney, E., apoptotic and necrotic death signaling and the role of caspases Westin, S., Thanos, D., Rosenfeld, M. G., Glass, C. K., and Collins, T. (1999) in determining the type of cell death, which may help under- Mol. Cell. Biol. 19, 6367– 6378 33. Hottiger, M. O., Felzien, L. K., and Nabel, G. J. (1998) EMBO J. 17, 3124 –3134 stand the general cell death mechanism. 34. Cheshire, J. L., and Baldwin, A. S., Jr. (1997) Mol. Cell. Biol. 17, 6746 – 6754 In conclusion, we report a novel signal transduction of IFNg/ 35. Vercammen, D., Brouckaert, G., Denecker, G., Van de Craen, M., Declercq, W., TNFa synergism in the induction of ME-180 cell apoptosis: Fiers, W., and Vandenabeele, P. (1998) J. Exp. Med. 188, 919 –930 36. Eguchi, Y., Shimizu, S., and Tsujimoto, Y. (1997) Cancer Res. 57, 1835–1840 37. Kane, D. J., Sarafian, T. A., Anton, R., Hahn, H., Gralla, E. B., Valentine, J. S., K. Suk, I. Chang, Y.-H. Kim, J. Y. Kim, and M.-S. Lee, unpublished Ord, T., and Bredesen, D. E. (1993) Science 262, 1274 –1277 data. 38. Cory, S. (1995) Annu. Rev. Immunol. 13, 513–543

Journal

Journal of Biological ChemistryUnpaywall

Published: Apr 1, 2001

There are no references for this article.