Cell Line Death Pathway in a Human Myelomonocyte -Mediated

Inhibition of p38 Kinase Reveals a TNF-α -Mediated, Caspase-Dependent, Apoptotic Death Pathway in a Human Myelomonocyte Cell Line This information is current as of October 1, 2021. Jishy Varghese, Souvik Chattopadhaya and Apurva Sarin J Immunol 2001; 166:6570-6577; ; doi: 10.4049/jimmunol.166.11.6570 http://www.jimmunol.org/content/166/11/6570 Downloaded from

References This article cites 44 articles, 32 of which you can access for free at: http://www.jimmunol.org/content/166/11/6570.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on October 1, 2021 *average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2001 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Inhibition of p38 Kinase Reveals a TNF-␣-Mediated, Caspase-Dependent, Apoptotic Death Pathway in a Human Myelomonocyte Cell Line1

Jishy Varghese, Souvik Chattopadhaya, and Apurva Sarin2

TNF-␣ transduces signals of survival or death via its two receptors, R1/p55/p60 and RII/p80/p75. The role of caspases as effectors of cell death is universally accepted, although caspase inhibitors may potentiate TNF cytotoxicity in some instances. In conditions when macromolecular synthesis is blocked, caspases are part of the machinery that executes TNF-triggered apoptotic death in U937, a human myelomonocyte cell line, and in the Jurkat T cell line. However, inhibition of p38 mitogen-activated protein kinase (p38 MAPK) triggered TNF cytotoxicity in U937 cells and murine splenic macrophages, but not the Jurkat cell line. TNF induced expression of the antiapoptotic protein c-IAP2 (cytoplasmic inhibitor of apoptosis protein 2), and was blocked in the presence of a p38 MAPK inhibitor, which also induced caspase-dependent, TNF-mediated apoptosis in U937 cells. Thus, inhibition of p38 Downloaded from MAPK resulted in the activation of caspase 9 and cleavage of the adaptor molecule BH3 interacting domain death agonist, and blocked NF-␬B-mediated transactivation, without affecting the nuclear translocation of NF-␬B. Collectively, these data show that activation of p38 MAPK is critical to cell survival by TNF in U937 cells, and demonstrate lineage-speciﬁc regulation of TNF- triggered signals of activation or apoptosis. The Journal of Immunology, 2001, 166: 6570–6577.

umor necrosis factor-␣ is a multifunctional cytokine that gen-activated protein kinase (MAPK) kinase family of protein ki- http://www.jimmunol.org/ transduces signals of survival, differentiation, and death nases (9). T via its receptors TNFRI/p55/p60 and TNFRII/p75/p80 in TNF also activates the p38 MAPK pathway (3, 10, 11). MAPK diverse cell types (1–3). TNF elicits diverse biological conse- family members are required for signal transduction in response to quences by inducing the expression of various genes, and in many many stimuli including cytokines, and p38 MAPK have been im- systems the expression of these is regulated by the transcription plicated in the transcription-activating potential of NF-␬B (12–14). factor, NF-␬B (4, 5). NF-␬B can regulate the expression of genes Transcription factors such as ATF-2 (activating transcription fac- involved in inflammation, cell cycle regulation, immune responses tor-2), Elk-1, etc., are also activated via p38 MAPK, which con- via the expression of cytokines, and antiapoptotic molecules that tributes to TNF-triggered gene expression (11, 15), and existing protect cells from the deleterious effects of TNF (5, 6). The p50/ data support a role for p38 MAPK activation in both death and by guest on October 1, 2021 p65 NF-␬B heterodimer normally exists in the cytosol, with the survival pathways (3, 16–19). DNA-binding dimer held as an inactive complex with the inhibitor In situations in which TNF induces death, the cytoplasmic do- subunit I-␬B3 (inhibitory protein that dissociates from NF-␬B) (7). mains of TNFR, primarily TNFRI, recruit the adapter proteins, Nuclear translocation and activation of NF-␬B occur on the phos- TNFR-associated death domain, and Fas-associated death domain, phorylation and subsequent degradation of I-␬B (8), which re- which in turn activate the initiator caspase, Fas-associated death leases NF-␬B to bind ␬B response elements and initiate transcrip- domain-like IL-1␤-converting enzyme/procaspase 8. Caspase 8 tion. Phosphorylation of I-␬B is achieved by sequential steps triggers apoptotic damage by activating multiple effector caspases, involving multisubunit kinase complexes that belong to the mito- each of which target a limited set of cellular proteins (2). Similarly, procaspase 9 also functions as an initiator/regulator caspase. The N-terminal prodomain of this caspase interacts with Apaf-1 in the presence of cytochrome c and dATP, resulting in the cleavage and National Centre for Biological Sciences, Bangalore, Karnataka, India activation of caspase 9, which activates caspases 3, 6, and 7. Thus, Received for publication October 13, 2000. Accepted for publication March 22, 2001. there are at least two major pathways, initiated by caspases 8 and The costs of publication of this article were defrayed in part by the payment of page 9, which can activate caspase cascades in cells. Effector caspases charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. are required for the execution phase of apoptosis and are respon- 1 This study was funded by a core grant to A.S. from National Centre for Biological sible for many of the morphological changes and cellular damage Sciences, Tata Institute of Fundamental Research, India. that occur during cell death. The essential role of these proteases 2 Address correspondence and reprint requests to Dr. Apurva Sarin, National Centre has been demonstrated in diverse systems (20). TNF-mediated ap- for Biological Sciences, University of Agricultural Sciences, Gandhi Krishi Vigyan optosis is also dependent on caspase activation (1, 2, 21), but par- Kendra Campus, New Bellary Road, Bangalore 560065, Karnataka, India. E-mail address: [email protected] adoxically, a number of studies have shown that inhibition of 3 Abbreviations used in this paper: I-␬B, inhibitory protein that dissociates from NF- caspase activity can lead to necrotic death in many cell types ␬B; ATF-2, activating transcription factor-2; BD-FMK, Boc-Asp(O-methyl)-FMK; (22–25). BID, BH3 interacting domain death agonist; c-IAP, cytoplasmic inhibitor of apoptosis Ј In this study, we identify a novel role for p38 MAPK in the protein; CHX, cycloheximide; DiOC6, 3,3 -diethyloxacarbocyanine; Hsp27, heat- shock protein 27; IETD-FMK, Ile-Glu-Thr-Asp-FMK; JC-1, 5,5Ј,6,6Ј-tetrachloro- TNF-triggered antiapoptotic response in U937 cells. We find that 1,1Ј,3,3Ј-tetraethyl-benzimidazoleyl-carbocyanine iodide; LEHD-FMK, Leu-Glu- p38 MAPK is necessary to TNF-triggered antiapoptotic gene ex- His-Asp-FMK; MAPK, mitogen-activated protein kinase; TRAF-2, TNFR-associated factor-2; VEID-FMK, Val-Glu-Ile-Asp-FMK; ZVAD-FMK, Z-Val-Ala-Asp(O-methyl)- pression, and inhibition of p38 MAPK triggers a TNF-mediated, fluoromethyl ketone. caspase-dependent apoptotic death pathway in U937 cells.

Materials and Methods volume of PBS, mounted on slides, and photographed using a Pro-Series 3, Cells and reagents CCD camera (Media Cybernetics). U937, a human myelomonocytic cell line, and Jurkat, a T lymphoblastoid Western blot analysis cell line of human origin, were used in all experiments. Murine splenic The 106 cells treated under various conditions were harvested, washed macrophages were isolated by adherence on plastic petri dishes. In each ϫ 6 twice by centrifugation in chilled PBS, lysed by the addition of SDS sam- experiment, unfractionated cells (20 10 cells/ml) from three individual ple buffer (62.5 mM Tris-HCl, 2% w/v SDS, 10% glycerol, 50 mM DTT, spleens were incubated for 45 min in complete medium in tissue culture and 0.1% bromphenol blue), vortexed to reduce sample viscosity, dena- dishes. At the end of incubation, floating cells were removed by gentle tured by boiling, and then cooled on ice. Samples were microcentrifuged swirling and decantation of dish contents. Plates were washed twice with for 5 min before loading on 10% SDS-PAGE gels. Proteins were trans- complete medium using a similar procedure. After the final wash, 2 ml of ferred onto Hybond-ECL nitrocellulose membrane (Amersham, Little cold medium was added to dishes, and adherent cells were removed using Chalfont U.K.). Membranes were blocked for1hinblocking buffer (5% a cell scraper and used in functional assays. rTNF and Abs to TNFR1 and Blotto, 0.1% Tween 20 in PBS) and incubated overnight with primary Ab TNFRII were obtained from R&D Systems (Minneapolis, MN). Cyclohex- in blocking buffer at 4°C, with gentle rocking. Membranes were washed imide (CHX), Hoechst 33342, and propidium iodide were obtained from three times for 5 min each with 10 ml of PBS containing 0.1% Tween-20, Sigma (St. Louis, MO). The p38 MAPK inhibitor PD 169316 was pur- and then incubated for1hat25°C with goat anti-rabbit HRP (Pierce, chased from Calbiochem (San Diego, CA). The peptide inhibitors Z-Val- Rockford, IL), diluted 1/5000 in blocking buffer. After another three Ala-Asp(O-methyl)-fluoromethyl ketone (ZVAD-FMK), Boc-Asp(O- washes, proteins were detected by chemiluminescence, according to the methyl)-FMK (BD-FMK), Ile-Glu-Thr-Asp-FMK (IETD-FMK), Val-Glu- manufacturer’s instructions (Pierce). Ile-Asp-FMK (VEID-FMK), Leu-Glu-His-Asp-FMK (LEHD-FMK), and Z-Phe-Ala-FMK were obtained from Enzyme Systems Products (Dublin, NF-␬B reporter assay CA), dissolved as 20 mM stock solutions in DMSO, and stored at Ϫ70°C. The 5,5Ј,6,6Ј-tetrachloro-1,1Ј,3,3Ј-tetraethyl-benzimidazoleyl-carbocya- A total of 6–10 ϫ 106 U937 cells was transfected with 5 ␮g of thymidine Ј ␬ ␮ Downloaded from nine iodide (JC-1) and 3,3 -diethyloxacarbocyanine (DiOC6) were ob- kinase-6 NF- B luciferase by electroporation at 250 mV and 960 F ca- tained from Molecular Probes (Eugene, OR), made up as stock solutions in pacitance. Twenty-four hours after transfection, live cells were harvested DMSO, and stored at Ϫ70°C. Abs to cleaved caspase 9, phosphoisoforms by Ficoll density centrifugation, and cells were treated with TNF (10 ng/ of p38, and ATF-2 were from New England Biolabs (Beverly, MA); cy- ml), or TNF ϩ PD, or continued in medium alone. Six to 8 h after initiation toplasmic inhibitor of apoptosis protein 1 (c-IAP1), c-IAP2, TNFR-asso- of treatment, cell lysates were assayed for luciferase activity using a sub- ␬ ciated factor-2 (TRAF-2), BclxL, p38 MAPK, Fas, and NF- B were from strate provided by Promega (Madison, WI), according to the manufactur- Santa Cruz Biotechnology (Santa Cruz, CA); and biotin- or avidin-conju- er’s instructions, and read using a luminometer from Turner Design (Palo

gated Abs were obtained from Jackson ImmunoResearch (West Grove, Alto, CA). Promoter activity is expressed as fold induction over readings http://www.jimmunol.org/ PA). Ab to BH3 interacting death domain agonist (BID) was obtained from obtained from mock-transfected cells cultured in medium alone. R&D Systems. Apoptotic nuclear damage and cell lysis Results Caspase inhibitors block TNF ϩ CHX-induced apoptosis in Cell lysis was quantitated by ﬂow cytometry using propidium iodide. Ap- U937 and Jurkat cells optotic nuclear morphology was assessed using Hoechst 33342, as described previously (26). All assays of nuclear morphology were scored U937 cells were cultured with 5 ng/ml TNF in the presence or double blind. For imaging nuclei, cells were ﬁxed with 0.1% paraformal- absence of a broad spectrum caspase inhibitor peptide, ZVAD- dehyde for 10 min at room temperature, then washed and stained with 20 FMK or CHX. Cells were scored for lysis using propidium iodide

nM of the dye SYTOX Green (Molecular Probes) for 10 min at ambient by guest on October 1, 2021 temperature. At the end of incubation, the cells were washed and resus- and apoptotic nuclear morphology using Hoechst 33342 after 18 h pended in PBS (106/ml) for analysis of nuclear morphology. in culture. As seen in Fig. 1A, U937 cultured with TNF alone showed minimal levels of apoptosis (Ͻ10% apoptotic nuclei at Mitochondrial transmembrane potential 18 h), but the presence of 200 ng/ml CHX resulted in a sharp

For DiOC6 staining, cell pellets were incubated for 20 min in the dark at increase in apoptotic nuclear damage in these cells (Fig. 1, A and 37°C in 0.5 ml staining solution (PBS containing 40 nm DiOC6), as described previously (26). Cells were washed in excess PBS and resuspended in 0.5 ml PBS containing propidium iodide to identify dead cells, and the propidium iodide-negative subset was analyzed flow cytometrically. For staining with JC-1, cell pellets were resuspended in staining solution (200 ␮l PBS containing 5 ␮g/ml JC-1) and incubated at 37°C for 10 min. At the end of incubation, cells were washed in excess PBS and resuspended in 0.3 ml PBS and analyzed flow cytometrically, according to the manufacturer’s instructions. All flow cytometric analysis was performed on a Becton Dick- inson FACS (Becton Dickinson, Mountain View, CA). Immunofluorescence analysis by flow cytometry Cells fixed with 1% paraformaldehyde at room temperature for 20 min were washed twice in chilled permeabilization buffer (PBS containing 0.3% saponin, 0.1% sodium azide, and 1% FCS). A total of 20 ␮lof relevant Ab (diluted in permeabilization buffer) was added directly to the cell pellets, and the cells were incubated on ice for 30 min. Primary Ab was followed by secondary biotin-conjugated Ab, and all samples were finally stained with PE-avidin. Each step of Ab staining was followed by two washes in permeabilization buffer to remove excess Ab. At the end of the incubation with PE-avidin, cells were washed once with permeabilization ϩ buffer, followed by a wash in FACS buffer (PBS containing 1% FCS and FIGURE 1. TNF CHX induces apoptotic death in U937 cells. A, ϫ 5 0.1% sodium azide), and resuspended in 0.4 ml FACS buffer for flow U937 (4 10 cells/ml) were cultured for 12 h in the presence of 5 ng/ml cytometric analysis. TNF or TNF and 25 ␮M ZVAD-FMK or TNF and 200 ng/ml CHX. Cells were assessed for DNA damage by apoptotic nuclear morphology (dark Immunofluorescence analysis by microscopy bars) or lysis (shaded bars) by flow cytometry, as described in Materials Cells were fixed in 70% ethanol for 2 h, washed twice in saponin buffer, and Methods. B and C, Photomicrograph of nuclear morphology of U937 and stained as described for flow cytometric analysis, except that biotin- cells, stained with the nuclear dye Sytox green after treatment with TNF ϩ conjugated Ab was followed by avidin conjugated to Alexa-546 (Molec- CHX (upper panel)orTNFϩ ZVAD (lower panel) under conditions de- ular Probes). After the last wash, cells were resuspended in a minimal scribed for A. 6572 p38 KINASE BLOCKS TNF-MEDIATED APOPTOSIS IN U937 CELLS

B). As previously reported (25), the caspase inhibitor ZVAD-FMK TNF ϩ CHX for 16 h. All the peptides tested, including LEHD- also potentiated TNF-induced death (Fig. 1A), but a large propor- FMK, blocked nuclear damage (Fig. 2C), although ZVAD-FMK tion of the dead (propidium iodide-positive) cells in the TNF ϩ expectedly was most potent in inhibiting apoptotic death at con- ZVAD preparations had a distorted, condensed morphology (Fig. centrations as low as 5 ␮M (data not shown) in these cells. To 1, A and C), and did not show distinct nuclear fragmentation. address the possibility of nonspecific effects of the peptides, we To explore the mechanism by which TNF ϩ CHX induced apo- also tested an inhibitor of cathepsin, Z-Phe-Ala-FMK, which did ptotic death, we used cell-permeable, peptide inhibitors to investigate not block TNF ϩ CHX-mediated death in these cells (data not the role of caspases in this pathway. Maximal levels of apoptotic shown). nuclear damage were induced in 6–8 h by TNF (5 ng/ml) ϩ CHX ϩ (500 ng/ml) in U937 cells (Fig. 2A). Under these conditions, pre- TNF CHX triggers the activation of caspase 9 in Jurkat cells treatment with BD-FMK, a broad spectrum caspase inhibitor; The differential induction of caspase 9 in Jurkat cells, suggested in VEID-FMK, an inhibitor of caspase 6; or IETD-FMK, an inhibitor the previous experiments, was confirmed by assaying for the ex- of caspase 8, blocked TNF ϩ CHX-induced apoptotic nuclear pression of cleaved caspase 9 in intact, treated cells. Furthermore, damage in a dose-dependent manner (Fig. 2A). In the same exper- to aid comparison and normalize the kinetics of the apoptotic re- iment, BD-FMK and VEID-FMK only partially inhibited the loss sponse in the two cells, we treated Jurkat cells with TNF in the of mitochondrial transmembrane potential (Fig. 2B), while IETD- presence of a high dose of CHX (5 ␮g/ml) so as to induce TNF ϩ FMK had no effect, even at concentrations that block nuclear dam- CHX-mediated apoptosis with kinetics similar to U937 cells. Us- age. Concentrations of IETD-FMK higher than 50 ␮M have not ing a polyclonal Ab that specifically recognized the cleaved form been tested, as the resultant toxicity masked the protective effects, of the endogenous protease, Jurkat cells treated with the higher if any, of this inhibitor. In contrast, LEHD-FMK, an inhibitor of dose of CHX and TNF were stained after 2 or 4 h for expression Downloaded from caspase 9, did not block either nuclear or mitochondrial damage in of caspase 9. A small shift indicative of cleaved caspase 9 was these cells (Fig. 2, A and B). The partial inhibition of mitochondrial detected at 2 h (data not shown); however, by 4 h all the cells damage by the caspase inhibitors (Fig. 2B) suggests a role for other stained positive with this Ab (filled histogram, Fig. 2D). Specific proapoptotic molecules such as reactive oxygen species in this fluorescence was not detected in control, unstimulated cells, or in pathway (22, 25). ZVAD-FMK expectedly did not block either cells treated with Ab isotype controls (dotted and shaded line open nuclear (Fig. 2A) or mitochondrial damage (Fig. 2B) triggered by histograms, Fig. 2D). Under these conditions, TNF ϩ CHX in- http://www.jimmunol.org/ TNF ϩ CHX. The peptides used as caspase inhibitors are based on duced 60.5% apoptotic nuclear damage by 6 h, which was reduced optimal peptide recognition motifs that have been described for the to 20.6% in LEHD-FMK-treated cells. As expected from the func- different caspases (27). These have been used to design caspase- tional assays, cleaved caspase 9 could not be detected in U937 specific substrates as well (28). Although both BD-FMK and cells (Fig. 2E) at either early or later time points in the death ZVAD-FMK are broad spectrum inhibitors of caspases, differ- pathway, although U937 treated with TNF ϩ CHX (0.5 ␮g/ml) ences in their ability to inhibit death have been observed in other showed a measurable increase in nuclear damage by 2 h. Thus, systems as well (29). A recent study has suggested that ZVAD- TNF ϩ CHX induced 73% apoptotic damage compared with 2.3% FMK acts at a very early step in apoptotic cascades to block apoptotic nuclei scored in control, untreated cells 6 h after initia- by guest on October 1, 2021 caspases or a hitherto undefined, but related activity in cells (28). tion of the assay. These data indicate a role for caspase 9 in TNF- Similar assays performed in the Jurkat T cell line showed that mediated apoptosis in Jurkat cells, but not U937 cells, in condi- Jurkat cells are relatively resistant to TNF ϩ CHX compared with tions when TNF cytotoxicity is induced by CHX. U937, in that maximal levels of apoptotic nuclear damage are only Although CHX has been used routinely in many studies to an- observed 16–18 h after culture. Jurkat cells were pretreated with alyze TNF-mediated cytotoxicity, it can affect multiple events in 25 ␮M of the various peptides for 45 min and then cultured with cells. Notwithstanding, the experiments with CHX in this study

FIGURE 2. Caspase inhibitors block TNF ϩ CHX- mediated apoptosis in U937 and Jurkat cells. A and B, U937 cells (4 ϫ 105 cells/ml) were pretreated with different concentrations of peptides for 45 min and then cultured with indicated concentrations of TNF and CHX for 8 h. At the end of culture, cells were harvested and stained with Hoechst 33342 to assess apoptotic nuclear morphology (A), or with the dye JC-1 (B) to assess mitochondrial integrity, as described in Materials and Methods. C, Jurkat cells were set up under similar conditions as in A, except that cells were harvested after 18 h and assayed for apoptotic nuclei. Figures show values that are mean of three independent experiments. D and E, Expression of cleaved caspase 9 in Jurkat and U937 cells. D, Jurkat cells were treated with CHX (5 ␮g/ml) and TNF (5 ng/ml) and stained for cytoplasmic expression of caspase nine, 4 h post treatment. Control, unstimulated cells and cells stained with isotype Ab gave similar levels of staining. Cleaved caspase 9 is indicated by the ﬁlled histogram. E, U937 cells were treated with CHX (0.5 ␮g/ml) and TNF (5 ng/ml) and stained for caspase nine at 4 h. Caspase 9 was not detected at any time point tested. The experiment in D and E is representative of two separate experiments. The Journal of Immunology 6573

suggest the contextual regulation of TNF-initiated signaling events possible effector of Fas-mediated death has been recently sug- in the two cell types under study. In U937 cells, TNF induced gested in Jurkat cells (33). In subsequent experiments, we have necrosis in the presence of a caspase inhibitor (Fig. 1 and Ref. 25), further characterized the TNF ϩ PD-mediated death pathway in and activates proapoptotic caspases in the presence of CHX (Fig. U937 cells. 1, B and D, and Fig. 2). Thus, it was of interest to try and identify the nature of the input signals that can regulate caspase activation in TNF-treated U937 cells Inhibition of p38 MAPK triggers a caspase-dependent death pathway in U937 cells Inhibition of p38 MAPK potentiates TNF cytotoxicity in U937 cells Because PD169316 triggered TNF-mediated apoptosis, we tested the effect of caspase inhibitors on TNF ϩ PD-mediated apoptosis TNF signaling leads to the activation of kinase signaling cascades, in the U937 cell line. In the experiment shown in Fig. 3E,25␮M and these can regulate caspase activation in various cell death of the caspase inhibitors BD-FMK, VEID-FMK, or IETD-FMK pathways. To further characterize the regulation of TNF-mediated blocked TNF ϩ PD-mediated apoptosis, whereas ZVAD-FMK signaling in U937 cells, we tested for the effect of an inhibitor of was unable to block apoptotic nuclear damage (Fig. 3E). Unex- p38 MAPK, PD169316 (30–32). As shown in Fig. 3A, TNF, in the pectedly, LEHD-FMK, the inhibitor of caspase 9, also blocked presence of PD169316 (TNF ϩ PD), induced apoptotic nuclear nuclear damage in TNF ϩ PD-treated cells. damage at concentrations of PD that were nontoxic (Fig. 3A, open These experiments suggested a role for p38 MAPK in the mod- squares) to the cells. Agonist Ab to TNFR1 (Fig. 3B) triggered ulation of caspase 9 activation. Alternatively, it is also possible apoptotic death in the presence of 20 ␮M PD in a dose-dependent that caspase 9 activation or activity is directly inhibited by CHX, Downloaded from manner, but there was no induction of cytotoxicity by an Ab to TNFRII (Fig. 3B). These results are comparable with the effects of which may explain the lack of effect of the peptide LEHD-FMK on ϩ TNFR1 and TNFRII Abs in the presence of CHX (Fig. 3C) in these TNF CHX-mediated death (Fig. 2). Therefore, we tested for ϩ cells. PD169316 did not potentiate TNF-mediated apoptotic death inhibition of caspase 9 by CHX. As shown in Fig. 4A, TNF in Jurkat cells (Fig. 3D), suggesting a speciﬁc role for p38 MAPK PD-induced expression of activated caspase 9 (Fig. 4B, solid line) ␮ in TNF-mediated survival in U937 cells. was not blocked by 0.5 g/ml CHX (ﬁlled histogram in Fig. 4, A The immunoblot in Fig. 3F shows an experiment that demon- or B) or 50 ng/ml CHX (data not shown). Caspase 9 expression in http://www.jimmunol.org/ strates the induction of phosphorylated p38 MAPK in cells treated control, untreated cells is indicated by the solid line histogram in with 10 ng/ml TNF for 30 min (lane 2) as compared with untreated Fig. 4A. In functional assays for caspase 9 activity, we tested cells (lane 1). This experiment shows that pretreatment of cells whether 0.5 ␮g/ml (CHX) or lower concentrations of CHX (50 with PD169316 before the addition of TNF blocks the phosphor- ng/ml, CHXlo) would block the inhibitory effect of LEHD-FMK ylation of p38 MAPK (lane 3). Lane 4 served as a positive control, on TNF ϩ PD-mediated death. As shown in Fig. 4C, PD and CHX and shows the induction of phosphorylated p38 MAPK within 30 have additive effects on TNF-mediated death. Apoptotic death min in U937 cells treated with hydrogen peroxide (19). Interest- cells treated with LEHD-FMK ϩ TNF ϩ PD ϩ CHX were comparable with that of LEHD-FMK ϩ TNF ϩ CHX for each con-

ingly, in keeping with the differences in TNF-triggered signaling by guest on October 1, 2021 observed in U937 and Jurkat cells, a role for p38 MAPK as a centration of CHX. These data are consistent with the possibility

FIGURE 3. Effect of PD169316 on TNF-mediated apoptosis. A and B, U937 cells were cultured in A, without (Ⅺ) or with 5 ng/ml TNF (f) and increasing concentrations of PD169316, and in B, without (Ⅺ, E)or with (f, F)20␮M PD169316 and indicated concentrations of agonist Ab to TNFRI (f, Ⅺ) or TNFRII (F, E). C, U937 cells were cultured without (Ⅺ, E) or with (f, F) 500 ng/ml CHX and indicated concentrations of agonist Ab to TNFRI (f, Ⅺ) or TNFRII (F, E). D, Jurkat cells were cultured with 20 ␮M PD169316 in the presence (f) or the absence (Ⅺ) of increasing concentrations of TNF for 18 h. E, U937 cells were cultured with TNF (5 ng/ml) and 20 ␮M PD169316 for 18 h in the presence of 25 ␮M peptide-FMK, where indicated. In all experiments shown, apoptosis was assessed by apoptotic nuclear morphology. In A–C, a mean of ﬁve independent experiments has been plotted, and E is a mean of three experiments. F, Western blot analysis of phospho-p38 MAPK in U937. Cells were incubated for 2 h in 2% FCS in complete medium. Following this, cells were cultured for 30 min in medium alone (lane 1) or with 5 ng/ml TNF (lane 2), 5 ng/ml TNF after pretreatment with PD169316 (lane 3), or 0.2 mM hydrogen peroxide (lane 4), and assessed for the phosphorylation of p38 MAPK, as described in Materials and Methods. 6574 p38 KINASE BLOCKS TNF-MEDIATED APOPTOSIS IN U937 CELLS Downloaded from

FIGURE 4. Effect of CHX and IETD-FMK on TNF ϩ PD-induced caspase 9 activation. A and B, Expression of cleaved, caspase 9 (solid line) in U937 http://www.jimmunol.org/ cells treated with TNF (B) or left unstimulated (A) for 4 h. In the same experiment, shaded histograms represent the levels of cleaved caspase 9 in cells treated with 0.5 ␮g/ml CHX. C, U937 cells were cultured with TNF ϩ PD, TNF ϩ 0.5 ␮g/ml CHX (TNF ϩ CHX), TNF ϩ 50 ng/ml CHX (TNF ϩ CHXlo), or TNF ϩ PD and either 0.5 ␮g/ml CHX (TNF ϩ PD ϩ CHX) or 50 ng/ml CHX (TNF ϩ PD ϩ CHXlo). All experimental groups were cultured in the presence (ﬁlled histogram) or absence (speckled histogram) of 20 ␮M LEHD-FMK, and assayed for apoptotic nuclear damage at end of incubation. A representative experiment of two experiments is shown. In both experiments, PD alone, CHX alone, or PD ϩ CHX had no additional cytotoxicity over the control group (med) of cells (data not shown). D, U937 cells treated for 2 or 4 h with TNF or TNF ϩ PD were analyzed for levels of BID expression (ﬁrst panel). The same blots were stripped and reprobed for expression of Fas (second panel). In another set of experiments, cells were pretreated with 20 ␮M IETD-FMK (lane 3) and cultured with TNF ϩ PD (lane 2) and assayed for expression of BID. The same blot was again reprobed for Fas expression to

confirm that the loss of BID was not due to lower amounts of total protein loaded in the lanes. Jurkat cells treated with anti-Fas were run as controls in by guest on October 1, 2021 the same experiment. E–G, Cells treated with TNF (E)orTNFϩ PD (F) or IETD-FMK and TNF ϩ PD (G) were stained for cleaved, activated caspase 9 flow cytometrically. Values in the upper quadrant indicate the number of cells positive for cleaved caspase 9. that LEHD-FMK blocks the TNF ϩ PD-triggered, caspase 9 com- rophages. As shown in Fig. 5A, adherent macrophages isolated ponent of the death pathway and does not affect TNF ϩ CHX- from murine spleens when cultured overnight showed minimal in- mediated cell death. Additionally, these experiments suggest that duction of apoptotic damage. Cells treated with TNF for 18 h also caspase 9 activation is not regulated directly by CHX. showed minimal cytotoxicity as compared with control, untreated In preliminary experiments, shown in the immunoblot in Fig. cells. However, as was seen with U937 cells, PD169316 induced 4D, there is a loss of expression or immunoreactivity of the death- a striking increase in TNF-mediated apoptotic nuclear damage adaptor molecule BID, as early as 2 h following treatment with (solid bars) and cell lysis (hatched bars) as well. TNF ϩ PD (lanes 3 and 5) as compared with cells treated with We then tested whether p38 MAPK mediated cell survival by TNF (lane 2) or left untreated (lane 1). These conditions lead to modulating expression of various endogenous antiapoptotic pro- induction of apoptotic damage by 6 h. Furthermore, cells treated teins of the Bcl-2 family, including Bcl-2 and BclxL or the inhib- with IETD-FMK (lane 3) at concentrations that block cell death itors of apoptosis proteins, c-IAP1 and c-IAP2. The c-IAP family show restoration of BID expression as compared with cells treated are key players in the TNF-mediated antiapoptotic response (2, 6). with TNF ϩ PD (lane 2). All blots were reprobed for expression These proteins are transcriptionally regulated and block death by of Fas, which showed that changes in BID immunoreactivity were inhibiting the activation of caspases. Earlier experiments have not owing to alterations in cell number. Thus caspase 8-mediated shown that the expression of inhibitor of apoptosis protein is cleavage of BID can lead to the activation of caspase 9 in TNF ϩ among the early responses triggered by TNF via NF-␬B (6), and is PD-treated cells. As shown in Fig. 4E–G, IETD-FMK partially therefore sensitive to inhibition by agents such as CHX. U937 cells blocks the activation of caspase 9 (G) induced by TNF ϩ PD (F), were cultured with TNF for2hinthepresence or absence of 20 suggesting that caspase 9 activation may, in part, be dependent on ␮M PD169316, and analyzed for the cytoplasmic levels of c-IAP1, caspase 8-mediated cleavage of BID in U937 cells treated with c-IAP2, BclxL, Bcl-2, and TRAF-2. U937 cells express very low TNF ϩ PD. levels of c-IAP1 and c-IAP2 (data not shown). TNF-induced expression of c-IAP2 (Fig. 5C, solid line) was markedly reduced by Inhibition of p38 MAPK blocks antiapoptotic gene expression PD169316 (Fig. 5C, filled histogram); expression of c-IAP1 (Fig. To assess whether the effect of p38 MAPK on TNF signaling was 5B, solid line) was partially blocked (Fig. 5B, filled histogram), but limited to U937 cells alone, we also tested for the effect of PD on the high levels of constitutively expressed TRAF-2 (Fig. 5D, TNF-induced cytotoxicity in freshly isolated murine splenic mac- shaded histogram) were unchanged by this inhibitor (Fig. 5D). The Journal of Immunology 6575 Downloaded from http://www.jimmunol.org/

FIGURE 5. PD169316 potentiates TNF cytotoxicity in splenic macrophages and inhibits the expression of TNF-triggered antiapoptotic proteins in U937 cells, but does not inhibit NF-␬B nuclear translocation. A, Murine splenic macrophages were cultured overnight in the different conditions shown in the figure and analyzed for the induction of apoptotic nuclear damage (f) and cell lysis ( ). B–E, Expression of c-IAP1 (B) and c-IAP2 (C) in U937 cells treated with TNF (solid line, open histogram) or TNF ϩ PD (filled histogram), 2.5 h after addition of 5 ng/ml TNF. Cells stained with TRAF-2 (filled histogram)

are shown in D for both TNF and TNF ϩ PD-treated cells. The isotype controls for TNF or TNF ϩ PD-treated cells are indicated by the ﬁne or dotted by guest on October 1, 2021

line histogram, respectively, in each figure. U937 cells were stained for expression of BclxL (E) either in untreated cells (filled histogram, med) or after treatment with 5 ng/ml TNF (gray thick line, TNF) for 3 h. The dotted line and dark line represent isotype controls for untreated and TNF-treated cells, respectively. F–H, U937 cells were incubated for 20 min in the presence (G and H) or absence (F) of 5 ng/ml TNF. Cells in H were preincubated with 20 ␮M PD169316 for 30 min before the addition of TNF. Cells were harvested, fixed, and stained with an Ab that recognizes the p65 subunit of NF-␬B, as described in Materials and Methods. Cells were visualized using appropriate filters on an Olympus microscope and imaged using a CCD camera at ϫ40 magnification. I, Luciferase assay for NF-␬B promoter activity. U937 cells were transfected with the thymidine kinase-6 NF-␬B luciferase (luc) reporter plasmid or mock transfected, and 24 h later the transfection groups were treated with TNF or TNF ϩ PD. Eight hours after addition of TNF, cells were harvested, lysed, and assayed for NF-␬B-driven luciferase activity. Luciferase activity has been shown as fold induction over mock-transfected cells cultured in medium alone. J, U937 cells treated with TNF, TNF ϩ PD, and 0.2 mM hydrogen peroxide, or left untreated were analyzed for expression of the phosphoisoform of p38 MAPK (p-p38), and the same blot was stripped and reprobed for expression of the phosphoisoform of ATF-2 (p-ATF2) and total p38 MAPK.

␬ U937 cells do not express Bcl-2 (data not shown). BclxL is ex- blocked the nuclear translocation of NF- B. Thus, cells treated pressed in U937 cells at very low levels (Fig. 5E, filled histogram); with TNF show an early nuclear translocation (within 20 min of its expression is not enhanced further by TNF (Fig. 5E, heavy gray addition of TNF) of NF-␬B, which can be visualized by staining line), nor altered in TNF ϩ PD-treated cells. The role of p38 with an Ab that recognizes the p65 subunit of NF-␬B (Fig. 5, MAPK in TNF-mediated up-regulation of c-IAP2 in U937 cells F–H). As shown in Fig. 5F, the fluorescent signal with NF-␬BAb appears similar to its role in CD40-mediated signaling in dendritic in control, untreated cells is confined principally to the cytoplasmic cells (34). region of the cells. The nucleus can be seen as a central, sparsely stained region that is largely devoid of fluorescence in these cells. ␬ p38 MAPK does not affect the nuclear translocation of NF- B Exposure to TNF for 20 min results in the translocation of NF-␬B Activation of the transcription factor NF-␬B is central to the sur- to the nucleus, which can be visualized as the bright fluorescence vival pathway triggered by TNF in many systems. p38 MAPK in the nucleus (Fig. 5G). The fluorescence is sustained for at least appeared to function as a TNF-triggered survival signal in U937 45 min after addition of TNF (data not shown). The translocation cells, and because the effect of p38 kinase on NF-␬B-induced gene of NF-␬B is not inhibited in TNF ϩ PD-treated cells, as these cells expression has been shown in other systems (13, 14), we tested also show bright nuclear localized fluorescence (Fig. 5H), compa- whether p38 MAPK can also modulate NF-␬B activation. NF-␬B rable with that seen in cells treated with TNF alone. is normally trapped in the cytoplasm by its inhibitor I-␬B, but the To examine the effect of PD on TNF-induced NF-␬B-dependent association is regulated by phosphorylation of I-␬B by members of gene transcription, we transfected a NF-␬B reporter construct in the MAPK kinase family. We wished to test whether p38 MAPK U937 cells. As shown in Fig. 5I, U937 cells treated with TNF for 6576 p38 KINASE BLOCKS TNF-MEDIATED APOPTOSIS IN U937 CELLS

6–8 h in three experiments showed a 10- to 15-fold increase in lation of the transcription factor ATF-2 is inhibited by PD169316 NF-␬B reporter activity relative to control, untreated cells. This (Fig. 5J). c-IAP2, which has been functionally implicated in the induction was markedly blocked in cells treated with TNF ϩ PD, inhibition of TNF-mediated death, is transcriptionally regulated by but complete inhibition was not consistently observed in these as- NF-␬B (40). Transcription factors such as ATF-2 provide an ad- says. As shown in the immunoblots in panel J, concentration of PD ditional layer of regulation wherein p38 MAPK may modulate that results in a reduction in levels of phospho-p38 kinase in TNF ϩ NF-␬B-mediated transcription of genes in response to TNF. PD-treated cells also blocked induction of expression of the down- An earlier study has shown that inhibition of p38 MAPK does stream transcription factor phospho-ATF-2 in these cells. Hydro- not potentiate TNF cytotoxicity in L929 cells (38). We attribute a gen peroxide (H2O2)-treated cells were used as a positive control role in cell survival to p38 MAPK, because inhibition of p38 in these experiments, as described for Fig. 3. MAPK triggered death in U937 and splenic adherent cells (Figs. 3 and 5). This study shows that inhibition of p38 MAPK resulted in Discussion activation of caspase 9 and the loss of full-length BID. Addition- TNF triggers the expression of a largely overlapping battery of ally, caspase 8-dependent cleavage of the adaptor molecule BID genes in diverse cell types (2, 6), although it elicits widely varying and partial inhibition of caspase 9 activation by an inhibitor of responses of survival, death, or differentiation. This suggests that caspase 8 suggest that caspase 9 may, in part, be activated by a additional cell-specific factors, not necessarily transcriptionally caspase 8-dependent signaling event. regulated, modify the outcome of TNF signaling, and this study Procaspase 9 is recruited to an apoptosome complex that com- identifies p38 MAPK as one such candidate in U937 cells. Al- prises several Apaf-1 molecules, the oligomerization of which is though other signaling pathways have not been examined in this dependent on cytochrome c and dATP. We do not detect an early study, we find that p38 MAPK is a critical component of the TNF release of cytochrome c in these cells (data not shown). Heat-shock Downloaded from survival pathway in U937, a myelomonocyte cell line, and in mu- protein (Hsp) 27, a small molecular mass protein that is a pleio- rine splenic macrophage preparations as well. In T cells, p38 tropic inhibitor of cell death, is another negative regulator of MAPK is believed to contribute to cell death triggered by Fas (33), caspase 9 activation that binds cytochrome c and inhibits forma- although a more recent study has shown that TNF-triggered acti- tion of the apoptosome complex in cells (41). Hsp27 can also bind vation of p38 MAPK in Jurkat cells may not contribute to the Akt kinase/protein kinase B, which also has the capability of phos- apoptotic response in these cells (35). Thus, p38 kinase activity phorylating and inactivating caspase 9 (42). Hsp27 is phosphory- http://www.jimmunol.org/ may be a critical determinant for TNF-triggered survival in some lated by p38 MAPK (43, 44), and it is possible that in the TNF- cells particularly of the myeloid lineage, including dendritic cells mediated survival pathway, activated p38 MAPK phosphorylates and eosinophils (12, 18), but may also contribute to death in other Hsp27, which inactivates the apoptosome and contributes to the cells (16, 17). inhibition of caspase 9 activation in U937 cells. p38 MAPK has been implicated as a mediator of apoptosis (16, Based on biochemical and functional assays, this study provides 17), but recent reports have also described a protective role for this three lines of evidence that p38 MAPK is a component of the kinase in TNF-mediated signaling (3, 19). A recent report has in- survival response triggered by TNF in U937, a myelomonocyte dicated a protective role for p38 MAPK in response to oxidative cell line. Inhibition of p38 MAPK triggers TNF cytotoxicity in by guest on October 1, 2021 stress in U937 (19), although the role of antiapoptotic proteins, if functional assays of death in two experimental systems, the U937 any, in this system was not discussed. TNF-mediated death is cell line and splenic macrophages. Inhibition of p38 MAPK inhib- blocked by cytokines in T cells (36), and because p38 MAPK its NF-␬B-mediated gene activation in reporter assays, although regulates expression of cytokines in different cells (10, 12, 13), an nuclear translocation of NF-␬B is not blocked (Fig. 5). NF-␬Bis analysis of cytokines produced in response to TNF signaling in a well-defined mediator of the TNF-triggered survival response, U937 cells may reveal additional antiapoptotic signals generated and among other proteins regulates expression of the antiapoptotic via this pathway. The up-regulation of antiapoptotic proteins, such protein c-IAP2 (6, 39). The TNF-induced up-regulation of c-IAP2 as c-IAP2, is one way by which p38 MAPK contributes to TNF- is also blocked by the inhibitor of p38 MAPK (Fig. 5). Thus, the mediated survival, and thereby the determination of cell fate. experiments show that p38 MAPK can regulate NF-␬B-mediated We have attempted to address the possible interaction of TNF- gene transcription and also the expression of the antiapoptotic pro- triggered p38 MAPK and NF-␬B in this system. In cardiac myo- tein c-IAP2, although it remains to be established whether these cytes, p38 MAPK regulates the expression of IL-6, via interactions two events are related in the present system. between MKK6 (MAPK kinase), which lies directly upstream of Cells use several mechanisms to regulate the activity of the p38, and I-␬B kinase, which results in NF-␬B activation (13) in caspase family of proteases. Activation of caspase-9 may be reg- TNF-treated cells. This model of a direct impact of p38 MAPK on ulated by both transcription-dependent (c-IAP-mediated) (2, 6), NF-␬B activation is not supported by our data, which are more and transcription-independent (phosphorylation of regulatory pro- consistent with the effect of p38 MAPK on the TNF-triggered tran- teins) (41, 42) events. Although p38 MAPK may principally act scription-activating potential of NF-␬B. TNF-induced nuclear via a phosphorylation-dependent event, in this study we present translocation of NF-␬B (Fig. 5G) was not modulated by PD169316 evidence that this kinase can also intersect a key transcription- (Fig. 5H), but affected NF-␬B transactivation (Fig. 5I) and expres- dependent pathway and regulate TNF-mediated death in cells. sion of the antiapoptotic gene inhibitor of apoptosis protein-2 (Fig. Thus, these experiments provide additional evidence of multiple 5C) in these cells. A similar requirement for the p38 MAPK path- protective mechanisms influenced by this kinase. Collectively, the way in NF-␬B function, independent of its DNA binding, has been data indicate that the p38 MAPK pathway is critical to the TNF- reported in response to CD40 and TNF in other systems (14, 37, triggered antiapoptotic response and functions as a key cell fate 38). The mechanism by which p38 MAPK modulates NF-␬B tran- determinant in U937 cells. scription is not known. One possibility consistent with the experimental results in this study and others is that p38 MAPK or a Acknowledgments downstream kinase phosphorylates a transcriptional coactivator We thank Dr. Sudhir Krishna for the gift of the Ab to NF-␬B, the thymi- such as CBP/p300, which may modulate NF-␬B-driven gene ac- dine kinase-6 NF-␬B luciferase reporter construct, and V. Karthik for as- tivation (39). We also find that p38 MAPK-mediated phosphory- sistance with the reporter assays. We are grateful to Dr. Gaiti Hasan and V. The Journal of Immunology 6577

Karthik for their comments on the manuscript, and Khursheed Wani for his gene induction by a cycloheximide-sensitive factor occurs at the level of or up- assistance with preparation of the manuscript. stream of Fas-associated death domain protein. J. Biol. Chem. 275:24357. 24. Luschen, S., S. Ussat, G. Scherer, D. Kabelitz, and S. Adam-Klages. 2000. Sen- sitization to death receptor cytotoxicity by inhibition of FADD/caspase signal- References ling: requirement of cell cycle progression. J. Biol. Chem. 275:24670. 1. Ashkenazi, A., and V. M. Dixit. 1998. Death receptors: signalling and modula- 25. Khwaja, A., and L. Tatton. 1999. Resistance to the cytotoxic effects of tumor tion. Science 281:1305. necrosis factor ␣ can be overcome by inhibition of a FADD/caspase-dependent 2. Wallach, D., E. E. Varfolomeev, N. L. Malinin, Y. V. Golstev, A. V. Kovalenko, signaling pathway. J. Biol. Chem. 274:36817. and M. P. Boldin. 1999. Tumor necrosis factor and Fas signalling mechanisms. 26. Sarin, A., E. Hadid, and P. A. Henkart. 1998. Caspase dependence of target cell Annu. Rev. Immunol. 17:331. damage induced by cytotoxic T lymphocytes. J. Immunol. 161:2810. 3. Roulston, A., C. Reinhard, P. Amiri, and L. T. Williams. 1998. Early activation 27. Komoriya, A., B. Z. Packard, M. J. Brown, M.-L. Wu, and P. A. Henkart. 2000. of c-Jun N-terminal kinase and p38 kinase regulates cell survival in response to Assessment of caspase activities in intact apoptotic thymocytes using cell-per- tumor necrosis-factor ␣. J. Biol. Chem. 273:10232. meable fluorogenic caspase substrates. J. Exp. Med. 191:1819. 4. Wang, C.-Y., M. W. Mayo, and A. L. Baldwin, Jr. 1996. TNF- and cancer ther- 28. Thornberry, N. A., T. A. Rano, E. P. Peterson, D. M. Rasper, T. Timkey, apy-induced apoptosis: potentiation by inhibition of NF-␬B. Science 274:784. M. Garcia-Calvo, V. M. Houtzager, P. A. Nordstrom, S. Roy, J. P. Vaillancourt, 5. Barnes, P. K., and M. Karin. 1997. Nuclear factor-␬B: a pivotal transcription et al. 1997. A combinatorial approach defines specificities of members of the factor in chronic inflammatory disease. N. Engl. J. Med. 336:1066. caspase family and granzyme B: functional relationships established for key me- 6. Wang, C.-Y., M. W. Mayo, R. G. Korneluk., D. V. Goeddel, and A. L. Baldwin, diators of apoptosis. J. Biol. Chem. 272:17907. Jr. 1998. NF␬B antiapoptosis: induction of TRAF1 and TRAF2 and c-IAP1 and 29. Sarin, A., M.-L. Wu, and P. A. Henkart. 1996. Different ICE-family protease c-IAP2 to suppress caspase activation. Science 281:1680. requirements for the apoptotic death of T lymphocytes triggered by diverse stim- 7. Rice, N. R., and M. K. Ernst. 1993. In vivo control of NF-␬B activation by I␬B␣. uli. J. Exp. Med. 184:2445. EMBO J. 12:4685. 30. Kozawa, O., H. Kawamura, H. Matsuno, and T. Uematsu. 2000. p38 MAP kinase 8. Henkel, T., T. Machleidt, I. Alkalay, M. Kronke, Y. Ben-Nerriah, and is involved in the signalling of sphingosine in osteoblasts: sphingosine inhibits P. A. Baeuerle. 1993. Rapid proteolysis of I␬␤Ϫ␣ is necessary for activation of prostaglandin F(2␣)-induced phosphoinositide hydrolysis. Cell. Signal. 12:447. transcription factor NF-␬B. Nature 365:182. 31. Paine, E., R. Palmantier, S. K. Akiyama, K. Olden, and J. D. Roberts. 2000.

9. Malinin, N. L., M. P. Boldin, A. V. Kovalenko, and D. Wallach. 1997. MAP3K- Arachidonic acid activates mitogen-activated protein (MAP) kinase-activated Downloaded from related kinase involved in NF-␬B induction by TNF, CD95 and IL-1. Nature protein kinase 2 and mediates adhesion of a human breast carcinoma cell line to 385:540. collagen type IV through a p38 MAP kinase-dependent pathway. J. Biol. Chem. 10. Foey, A. D., S. L. Parry, L. M. Williams, M. Feldmann, B. M. Foxwell, and 275:11284. F. M. Brennan. 1998. Regulation of monocyte IL-10 synthesis by endogenous 32. Assefa, Z., A. Vantieghem, M. Garmyn, W. Declercq, P. Vandenabeele, IL-1 and TNF␣: role of p38 and p42/44 mitogen activated protein kinases. J. Im- J. R. Vandenheede, R. Bouillon, W. Merlevede, and P. Agostinis. 2000. p38 munol. 160:920. mitogen-activated protein kinase regulates a novel, caspase-independent pathway 11. Raingeaud, J., S. Gupta, J. S. Rogers, M. Dickens, J. Han, R. J. Ulevitch, and for the mitochondrial cytochrome c release in ultraviolet B radiation-induced R. J. Davis. 1995. Pro-inﬂammatory cytokines and environmental stress cause apoptosis. J. Biol. Chem. 275:21416. p38 mitogen-activated protein kinase activation by dual phosphorylation on 33. Wang, S., N. Nath, A. Minden, and S. Chellappan. 1999. Regulation of Rb and http://www.jimmunol.org/ serine and threonine. J. Biol. Chem. 270:7420. E2F by signal transduction cascades: divergent effects of JNK1 and p38 kinases. 12. Park, S.-M., H.-S. Kim, J. Choe, and T. H. Lee. 1999. Differential induction of EMBO J. 18:1559. cytokine genes and activation of mitogen-activated protein kinase family and 34. Aicher, A., G. L. Shu, D. Magaletti, T. Mulvania, A. Pezzutto, A. Craxton, and soluble CD-40 ligand and TNF in a human follicular dendritic cell line. J. Im- E. A. Clark. 1999. Differential role for p38 mitogen-activated kinase in regulating munol. 163:631. CD40-induced gene expression in dendritic cells and B cells. J. Immunol. 163: 13. Craig, R., A. Larkin, A. M. Martin, D. J. Thuerauf, C. Andrews, 5786. ␬ P. M. McDonough, and C. C. Glembotski. 2000. p38 MAPK and NF- B collab- 35. MacFarlane, M., G. M. Cohen, and M. Dickens. 2000. JNK (c-Jun-N-terminal orate to induce interleukin-6 gene expression and release: evidence for a cyto- kinase) and p38 activation in receptor mediated and chemically induced apoptosis protective autocrine signaling pathway in a cardiac myocyte model system. of T-cells: differential requirements for activation. Biochem. J. 348:93. J. Biol. Chem. 275:23814. 36. Sarin, A., M. Conan-Cibotti, and P. A. Henkart. 1995. Cytotoxic effects of TNF 14. Brent Carter, A., K. L. Knudston, M. M. Monick, and G. W. Hunnighake. 1999.

and lymphotoxin in T lymphoblasts. J. Immunol. 155:3716. by guest on October 1, 2021 The p38-mitogen-activated protein kinase is required for NF-␬B-dependent gene 37. Craxton, A., G. Shu, J. D. Graves, J. Saklatvala, E. G. Krebs, and E. A. Clark. expression. J. Biol. Chem. 274:30858. 1998. p38 MAPK is required for CD40-induced gene expression and proliferation 15. Martin-Blanco, E. 2000. p38 MAPK signalling cascades: ancient roles and new in B lymphocytes. J. Immunol. 161:3225. functions. BioEssays 22:637. 16. Aoshiba, K., S. Yasui, M. Hayashi, J. Tamaoki, and A. Nagai. 1999. Role of 38. Beyaert, R., A. Cuenda, W. V. Berghe, S. Plesance, J. C. Lee, P. Cohen, and p38-mitogen activated protein kinase in spontaneous apoptosis of human neutro- W. Fiers. 1996. The p38/RK mitogen-activated protein kinase regulates interleu- phils. J. Immunol. 162:1692. kin-6 synthesis in response to tumor necrosis factor. EMBO J. 15:1914. 17. Kummer, J. L., P. K. Rao, and K. A. Heidenreich. 1997. Apoptosis induced by 39. Gerritsen, M. E., A. J. Williams, A. S. Neish, S. Moore, Y. Shi, and T. Collins. withdrawal of trophic factors is mediated by p38 mitogen-activated protein ki- 1997. CREB-binding protein/p300 are transcriptional coactivators of p65. Proc. nase. J. Biol. Chem. 272:20490. Natl. Acad. Sci. USA 94:2927. 18. Tsukahara, K., A. Nakao, M. Hiraguri, S. Mike, M. Mamura, Y. Saito, and 40. Chu, Z. L., T. A. McKinsey, L. Liu, J. J. Gentry, M. H. Malim, and D. W. Ballard. I. Iwamoto. 1999. Tumor necrosis factor-␣ mediates antiapoptotic signals par- 1997. Suppression of tumor necrosis factor-induced cell death by inhibitor of ␬ tially via p38 MAP kinase activation in human eosinophils. Int. Arch. Allergy apoptosis c-IAP2 is under NF B control. Proc. Natl. Acad. Sci. USA 94:1007. Immunol. 120 (Suppl. 1):54. 41. Bruey, J.-M., C. Ducasse, P. Bonniaud, L. Ravagnant, S. A. Susin, 19. Kurata, S.-I. 2000. Selective activation of p38 MAPK cascade and mitotic arrest C. Diaz-Latoud, S. Gurbuxani, A.-P. Arrigo, G. Kroemer, E. Solary, and caused by low level of oxidative stress. J. Biol. Chem. 275:23413. C. Garrido. 2000. Hsp27 negatively regulates cell death by interacting with cy- 20. Los, M., S. Wesselborg, and K. Schultze-Osthoff. 1999. The role of caspases in tochrome c. Nat. Cell Biol. 2:645. development, immunity and apoptotic signal transduction: lessons from knockout 42. Cardone, M. H., N. Roy, H. R. Stennicke, G. S. Salvessen, T. E. Franke, mice. Immunity 10:629. E. Stanbridge, S. Frisch, and J. C. Reed. 1998. Regulation of cell death protease 21. Tiwari, M., and V. M. Dixit. 1995. Fas and tumor necrosis factor-induced apo- caspase 9 by phosphorylation. Science 282:1318. ptosis is inhibited by the poxvirus crmA gene product. J. Biol. Chem. 270:3255. 43. Hepburn, A., D. Demolle, J. M. Boeynaems, W. Fiers, and G. E. Dumon. 1988. 22. Vercammen, D., R. Beyaert, G. Denecker, V. Goossens, G. Van Loo, Rapid phosphorylation of a 27 kD protein induced by tumor necrosis factor. W. Declercq, J. Grooten, W. Fiers, and P. Vandenabeele. 1998. Inhibition of FEBS Lett. 227:175. caspases increases the sensitivity of L929 cells to necrosis mediated by tumor 44. Guesdon, F., N. Freshney, R. J. Waller, L. Rawlinson, and J. Saklatvala. 1993. necrosis factor. J. Exp. Med. 187:1477. Interleukin-1 and tumor necrosis factor stimulate two novel protein kinases that 23. Wajant, H., E. Haas, R. Schwenzer, F. Muhlenbeck, S. Kreuz, G. Schubert, phosphorylate the heat shock protein hsp27 and ␤-casein. J. Biol. Chem. M. Grell, C. Smith, and P. Scheurich. 2000. Inhibition of death receptor-mediated 268:4236.