of April 4, 2018.This information is current as

B ActivationκCaspase-8-Dependent NF-T Cell Activation through

+Cellular FLIP (Long Form) Regulates CD8

Q. Russell, Jurg Tschopp and Ralph C. BuddAustin Dohrman, Takao Kataoka, Solange Cuenin, Jennifer

http://www.jimmunol.org/content/174/9/5270doi: 10.4049/jimmunol.174.9.5270

2005; 174:5270-5278; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/174/9/5270.full#ref-list-1

, 19 of which you can access for free at: cites 35 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2005 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Cellular FLIP (Long Form) Regulates CD8� T Cell Activationthrough Caspase-8-Dependent NF-�B Activation1

Austin Dohrman,* Takao Kataoka,† Solange Cuenin,*‡ Jennifer Q. Russell,* Jurg Tschopp,‡

and Ralph C. Budd2*

Cellular FLIP long form (c-FLIPL) was originally identified as an inhibitor of Fas (CD95/Apo-1). Subsequently, additional func-tions of c-FLIPL were identified through its association with receptor-interacting protein (RIP)1 and TNFR-associated factor 2 toactivate NF-�B, as well as by its association with and activation of caspase-8. T cells from c-FLIPL-transgenic (Tg) mice manifesthyperproliferation upon activation, although it was not clear which of the various functions of c-FLIPL was involved. We havefurther explored the effect of c-FLIPL on CD8� effector T cell function and its mechanism of action. c-FLIPL-Tg CD8� T cells haveincreased proliferation and IL-2 responsiveness to cognate Ags as well as to low-affinity Ag variants, due to increased CD25expression. They also have a T cytotoxic 2 cytokine phenotype. c-FLIPL-Tg CD8� T cells manifest greater caspase activity andNF-�B activity upon activation. Both augmented proliferation and CD25 expression are blocked by caspase inhibition. c-FLIPL

itself is a substrate of the caspase activity in effector T cells, being cleaved to a p43FLIP form. p43FLIP more efficiently recruits RIP1than full-length c-FLIPL to activate NF-�B. c-FLIPL and RIP1 also coimmunoprecipitate with active caspase-8 in effector CD8�

T cells. Thus, one mechanism by which c-FLIPL influences effector T cell function is through its activation of caspase-8, which inturn cleaves c-FLIPL to allow RIP1 recruitment and NF-�B activation. This provides a partial explanation of why caspase activityis required to initiate proliferation of resting T cells. The Journal of Immunology, 2005, 174: 5270–5278.

T he potential role of death receptors or downstreamcaspases in cell processes such as growth, differentiation,or effector function, has received growing recognition in

a variety of cell types. Fas has been reported to be involved withhepatocyte regeneration (1), neurite dendrite outgrowth (2), fibro-blast proliferation (3), cardiac hypertrophy (4), and T cell costimu-lation (5). The concept that signaling molecules in the Fas pathwayare important to development is underscored by the finding thatmice genetically deficient for either Fas-associated death domainprotein (FADD),3 caspase-8, or cellular FLIP (c-FLIP), are all em-bryonically lethal due to a cardiac malformation (6). Furthermore,T cells from humans or mice lacking functional caspase-8 manifesta profound proliferation defect (7), consistent with earlier findingsthat caspase inhibitors block proliferation of human T cells (8, 9).Exactly how death receptors or associating caspases might pro-mote cell growth and differentiation is unknown, as are the poten-tial caspase target(s) for these processes.

We considered c-FLIP long form (c-FLIPL) as a caspase sub-strate that is potentially involved with cell growth. c-FLIPL is ahomolog of caspase-8 in possessing two death effector domains but

lacks a functional caspase domain due to a mutation of a criticalcysteine to tyrosine in the enzymatic domain (10). As such, c-FLIPL acts as a competitive inhibitor for recruitment of caspase-8to the death effector domain of FADD following Fas ligation (11).However, subsequent studies demonstrated the ability of c-FLIPL

to bind adaptor proteins that can link to the NF-�B and ERK path-ways. These adaptors include receptor-interacting protein (RIP)1,TNFR-associated factor (TRAF)2, and Raf-1 (12). Increased ex-pression of c-FLIPL in T cell lines augmented IL-2 production, andin transgenic mice, c-FLIPL enhanced T cell proliferation (13). Inaddition, c-FLIPL has a known caspase cleavage site at Asp376

resulting in caspase-8-dependent cleavage of full-length 55-kDac-FLIPL to p43FLIP (14–16). The functional significance ofc-FLIPL cleavage is unknown.

The current studies sought to define how increased c-FLIPL ex-pression results in enhanced T cell growth and whether cleavage ofc-FLIPL is required for this function. We observe that mice trans-genic for c-FLIPL in the T cell compartment have a decreasedactivation threshold to Ags bearing decreased TCR affinity and areless dependent on CD28 costimulation. The increased proliferationis due to augmented expression of CD25, consistent with knownincreased activation of ERK and NF-�B by c-FLIPL. These effectsof c-FLIPL are independent of Fas expression. Finally, the abilityof c-FLIP to recruit RIP1 is greatly increased after c-FLIPL iscleaved to p43FLIP. These findings suggest that, during T cell ac-tivation, c-FLIPL may be a critical caspase substrate that helpspromote the activation of NF-�B.

Materials and MethodsMice

c-FLIPL was expressed transgenically in T cell compartment as previouslyreported (13). Briefly, FLAG-tagged mouse FLIPL cDNA was inserted into atarget vector containing the �-globin promoter and a downstream human CD2locus enhancer element. Transgenic mice were screened by PCR of ear DNAusing the following primers: 5� primer, 5�-GGAGCCAGGGCTGGGCATAAAA-3�; and 3� primer, 5�-GACTCACCCTGAAGTTCTCAGGATCC-3�.

*Immunobiology Program, Department of Medicine, University of Vermont Collegeof Medicine, Burlington, VT 05405; †Center for Biological Resources and Informat-ics, Tokyo Institute of Technology, Yokohama, Japan; and ‡Institute of Biochemistry,University of Lausanne, Biomedical Research Center, Epalinges, Switzerland

Received for publication June 23, 2004. Accepted for publication February 15, 2005.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Institutes of Health (NIH) Grants AI36333and AI45666 (to R.C.B.). A.D. was supported by NIH Grant T32 CA09286.2 Address correspondence and reprint requests to Dr. Ralph C. Budd, ImmunobiologyProgram, University of Vermont College of Medicine, Given Medical Building, Bur-lington, VT 05405-0068. E-mail address: ralph.budd@uvm.edu3 Abbreviations used in this paper: FADD, Fas-associated death domain protein; c-FLIP, cellular FLIP; c-FLIPL, c-FLIP long form; v-FLIP, viral FLIP; RIP, receptor-interacting protein; TRAF, TNFR-associated factor; Tg, transgenic; OVAp, OVApeptide SIINFEKL; z-VAD-fmk, z-Val-Ala-Asp(OCH3)-fluoromethylketone.

The Journal of Immunology

Copyright © 2005 by The American Association of Immunologists, Inc. 0022-1767/05/$02.00

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Immunoblot using anti-FLIP mAb (Dave-2; Apoxis) further confirmedexpression of the transgene at levels 8- to 10-fold greater than wild-typec-FLIPL. The c-FLIPL-transgenic (Tg) mouse strain has been backcrossedto C57BL/6 mice (The Jackson Laboratory) for nine generations. Micewere maintained at the University of Vermont Animal Facility (AmericanAssociation for the Accreditation of Laboratory Animal Care approved),and experiments were conducted in accordance with Institutional AnimalCare and Use Committee-approved protocols.

OT-1 mice bear a transgenic TCR that recognizes chicken OVAp re-stricted to class I MHC, Kb, and were kindly provided by Drs. F. Carboneand M. Bevan (University of Washington, Seattle, WA) (17). OT-1 micewere maintained by breeding TCR transgenic male mice to normalC57BL/6 females. Offspring were screened for the clonotype TCR usinganti-V�2 mAb.

CD8� T cell purification and culture

Spleen cells after hemolysis by Gey’s solution were combined with lymphnode cells followed by negative selection to enrich for CD8� cells. Cellswere incubated with Abs to CD4 (GK1.5), MHC class II (3F12), NK1.1(PK136), and CD11b (all from BD Biosciences) for 30 min to remove,respectively, CD4� cells, B cells, NK cells, and macrophages. Sampleswere washed and then incubated with goat anti-rat/mouse IgG-labeledmagnetic beads (Qiagen) for 45 min followed by magnetic field separation.CD8� T cell purity was confirmed by the flow cytometry and was routinely�90%.

T cells were activated in culture medium (RPMI 1640 supplementedwith penicillin (200 �g/ml; Sigma-Aldrich), streptomycin (200 �g/ml; Sig-ma-Aldrich), glutamine (4 mM; Sigma-Aldrich), 2-ME (50 �M; Sigma-Aldrich), HEPES (10 mM; Sigma-Aldrich), and 8% FBS (Intergen)) usingplate-bound anti-CD3 (5 �g/ml; clone 145-2C11), anti-CD28 ascites (1:500), and recombinant human IL-2 (50 U/ml; Cetus), or OVAp or low-affinity variant G4 (SIIGFEKL) in the case of OT-I mice, for 2 days. Forproliferation studies, cell were pulsed at this time with [3H]thymidine foran additional 18 h before harvest. Cells propagated for longer periods wereremoved from anti-CD3-coated wells on day 2 and fed with fresh culturemedium containing IL-2.

Cytokine ELISA

Purified CD8� T cells were cultured at 106/ml with plastic-bound anti-CD3(5 �g/ml; 145-2C11) and soluble anti-CD28 (1/500 dilution of ascites;37.51) for 48 h, and 72 h. Quantification of cytokines (IL-2, IL-4, andIFN-�) in cell culture supernatants was performed using a sandwich ELISAas described (18).

RNA preparation and RNase protection assay

Total RNA was prepared from cultured CD8� T cells using Ultraspec(Biotecx) according to the manufacturer’s recommendation. CytokineRNA levels were determined by RNase protection assay using the Ribo-Quant multiprobe kit (BD Biosciences). Five micrograms of total RNA washybridized overnight with a 32P-labeled RNA probe, which had been syn-thesized from the multicytokine template set, after which free probe andother ssRNA were digested with RNase.

Abs and flow cytometry

Monoclonal anti-murine CD8� conjugated to Red613 was purchased fromInvitrogen Life Technologies. Monoclonal anti-murine CD4 conjugated toTriColor or PE was purchased from Caltag Laboratories. Monoclonal anti-murine V�2 conjugated to PE, monoclonal anti-murine CD69 conjugatedto PE, monoclonal anti-murine CD80 (B7.1) conjugated to FITC, mono-clonal anti-murine CD86 (B7.2) conjugated to PE, and monoclonal anti-murine H-2 Kb conjugated to biotin were purchased from BD Biosciences.

For flow cytometry, 750,000 cells were incubated in 0.1 ml of PBScontaining 0.5% BSA fraction V, 0.001% (w/v) sodium azide (PBS-azide)(Sigma-Aldrich), and the Abs listed above at 4°C for 30 min. After wash-ing with PBS-azide, cells were fixed in 1% methanol-free formaldehyde(Ted Pella) in PBS-azide. Samples were stored at 4°C until they wereanalyzed with a Coulter Elite flow cytometer calibrated using DNA checkbeads (Coulter).

AP-1-, NFAT-, and NF-�B-luciferase reporter mice andluciferase activity

AP-1-luciferase transgenic mice carry the luciferase gene driven by fourhuman collagenase 12-O-tetradecanoylphorbol-13-acetate-responsive ele-ments, which have high affinity for the AP-1 complex, in the context of therat minimal prolactin promoter, linked to the luciferase gene (19). The

NFAT-luciferase reporter mice bear three copies of the NFAT bindingsequence from the IL-2 gene linked to the luciferase gene (20). The NF-�B-luciferase reporter mice have the luciferase gene controlled by twocopies of �B sequences from the Ig� enhancer (21). Purified CD8� T cellswere activated with anti-CD3 (5 �g/ml) and anti-CD28 (1/500 dilution ofascites). At 48 h, and 72 h following activation, cells were harvested,washed with PBS, and lysed. The lysates were then analyzed using lucife-rin (Promega) and measured in a luminometer for 10 s. Four measurementswere made for each sample. Results are presented as the mean (�SEM)with background subtracted.

Caspase activity assay

Total cellular caspase activity was quantitated using DEVD-rhodamine(Promega) according to the manufacturer’s protocol. In brief, viable cellswere isolated using centrifugation over Lympholyte M (Cedarlane) andwere titrated as indicated in 100 �l of culture medium, and an equal vol-ume of DEVD-rhodamine was added to the cells according to the manu-facturer’s protocol. As the DEVD substrate is cleaved by caspases, rhoda-mine is released and measured by a fluorescent spectrophotometer at 2 h.

Western blot analysis

Cells were washed once in ice-cold PBS and solubilized in lysis buffer(0.5% Nonidet P-40, 50 mM Tris-HCl (pH 7.6), 150 mM NaCl, 10% glyc-erol, 2 mM DTT, protease inhibitor mixture (Complete; Boehringer Mann-heim). Postnuclear lysate proteins (40 �g per lane) were separated in12.5% SDS-PAGE. Proteins were transferred to polyvinylidene difluoridemembranes (Hybond-ECL; Amersham), and blots were blocked andprobed with the indicated Abs in 4% nonfat milk in TBS/0.1% Tween 20.Immunoreactive proteins were visualized using HRP-labeled conjugates(Jackson ImmunoResearch Laboratories) and ECL blotting substrate (Am-ersham Biosciences). Abs used were specific for caspase-8 and c-FLIP(Apoxis), and RIP1 (BD Biosciences).

Transfection studies

Human embryonic kidney 293 and 293T cells were seeded on 60-mmculture plates (3 ml/plate) the day before transfection, and transfected bythe calcium phosphate method with various expression vectors. The cellswere harvested 16 h after transfection and washed with PBS, and lysedwith lysis buffer. After repeated centrifugation, postnuclear lysates wereprecleared with Sepharose 6B for 1 h and then incubated with anti-FLAGM2 agarose (Sigma-Aldrich) for 3 h. Agarose beads were washed fourtimes with the lysis buffer. Immunoprecipitates and cell lysates were ana-lyzed by Western blotting. 293T cells were seeded on a 24-well cultureplate (0.6 ml/plate) the day before transfection, and transfected by thecalcium phosphate method with various expression vectors, together withthe NF-�B-driven luciferase reporter plasmid and the �-galactosidase plas-mid. Postnuclear lysates were subjected to the luciferase assay and �-ga-lactosidase assay, which was used to normalize transfection efficiency.

Biotin-VAD-fmk caspase precipitation assay

Viable T cells (freshly isolated or day 4 T cell blasts) were incubated with10 �M biotin-VAD-fmk (Enzyme System Products). Cell membranes weredisrupted using lysis buffer containing 20 �M biotin-VAD. Lysate (600�g) was then precleared by rocking with 40 �l of agarose beads (SantaCruz) at 4°C for 2 h. Supernatant was then incubated with 30 �l of strepta-vidin-Sepharose beads (Zymed) on a rocker at 4°C overnight. Beads werewashed five times in lysis buffer without Complete protease inhibitor.Beads were then boiled in loading buffer. Beads were removed by centrif-ugation, and immunoblot analysis was then performed on the supernatants.

ResultsCD8� T cells from c-FLIPL-Tg mice manifest increased CD28-independent proliferation to low-dose Ag peptide or low-affinitypeptide variant

Although caspase-8 has been demonstrated to be required for ac-tivation of both human and mouse T cells (7, 22), the process thatinitiates the caspase activity is not known, nor is the caspase-8substrate(s) that is responsible for promoting activation. We con-sidered the possibility that c-FLIPL might provide both of thesefunctions because it can both heterodimerize with and activatecaspase-8, and is also cleaved by caspase-8 (23). We have previ-ously observed that CD8� T cells from mice transgenic for c-

5271The Journal of Immunology

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

FLIPL (c-FLIPL-Tg) in the T cell compartment manifest hyper-proliferation upon activation (13). We sought to extend thesestudies to determine the mechanism of increased proliferation, andwhether caspase activity and c-FLIPL cleavage are in any waylinked to these functions. As previously reported, CD8� T cellsfrom c-FLIPL-Tg mice had increased proliferation in response tolow-dose CD3 stimulation (Ref. 13; Fig. 1A). At doses of anti-CD3of 5 �g/ml or higher, control and c-FLIPL-Tg CD8� T cells man-ifested similar rates of incorporation of [3H]thymidine. This re-flects the fact that, although cell cycling of c-FLIPL-Tg CD8� Tcells was still faster than wild-type CD8� cells even at higherconcentrations of anti-CD3 (see Fig. 4A below), increased celldeath of c-FLIPL-Tg CD8� T cells balanced this property (A.Dohrman, unpublished observations). Hyperproliferation was alsoapparent even in the absence of CD28 costimulation (Fig. 1B).

The augmented proliferation was also observed with Ag stim-ulation by crossing the c-FLIPL-Tg mouse to the OT-I mouse thatbears a transgenic TCR reactive to OVAp restricted to H-2Kb (17).This system allowed a better definition of the effects of decreasingthe MHC/peptide affinity for TCR. The OT-I TCR binds H-2Kb/SIINFEKL with an affinity of 6.5 �M, whereas the peptide variantSIIGFEKL (G4) binds with a lesser affinity of 10 �M (24). OT-1 � c-FLIPL-Tg T cells manifested increased proliferation toOVAp over a broad dose range (Fig. 1C). This was particularlypronounced with low-affinity G4, to which OT-I T cells prolifer-ated only minimally, whereas OT-1 � FLIP-Tg T cells had a con-siderably stronger response (Fig. 1D). The dramatic difference inthe proliferation intensity between OVAp and G4 reflects the non-linear relation between affinity of the TCR interaction and effectorfunction, as observed previously with OT-1 T cells (17). As notedin Fig. 1, A and C, at high doses of anti-CD3 (�5 �g/ml) or OVAp(�10�8 M), there was no difference in [3H]thymidine incorpora-tion. This most likely resulted from the increased cell cycling ofc-FLIPL-Tg T cells being balanced by increased death, due to in-creased caspase-8 activation by c-FLIPL (see Fig. 6).

Hyperproliferation of c-FLIPL-Tg CD8� T cells is due toincreased expression of CD25

IL-2 signaling represents one of the principal growth pathways forT cells. Production of IL-2 and expression of the high-affinity IL-2R�-chain (CD25) were thus examined on c-FLIPL-Tg CD8� Tcells. In parallel with the differences in proliferation, IL-2 produc-tion by c-FLIPL-Tg CD8� T cells was higher than from the equiv-alent population from normal littermate controls at low-dose anti-

CD3, but became equivalent at higher anti-CD3 concentrations(Fig. 2A). To determine whether this completely explained the dif-ference in proliferation, exogenous IL-2 was added to T cells ac-tivated by either low-dose anti-CD3 or with OT-I cells stimulatedwith G4 peptide. Although exogenous IL-2 increased the prolifer-ation of both populations, it remained higher in the c-FLIPL-TgCD8� T cells (Fig. 2, B and C).

The persistence of increased proliferation of c-FLIPL-Tg CD8�

T cells despite addition of exogenous IL-2 suggested that the aug-mented proliferation of c-FLIPL-Tg T cells was not due solely toa difference in IL-2 production, but resulted in part also from in-creased sensitivity to IL-2. This possibility was investigated fur-ther through an analysis of expression of CD25 and the activationmarker CD69 on c-FLIPL-Tg CD8� T cells following TCR stim-ulation. As shown in Fig. 3, following activation of purified T cellsvia CD3/CD28, these activation markers were more prominently

FIGURE 1. Increased proliferation of c-FLIPL-TgCD8� T cells to low-dose anti-CD3, Ag peptide, andlow-affinity peptide variant. A and B, Purified CD8� Tcells (5 � 104/well) from c-FLIPL-Tg mice or normallittermate controls (NLC) were activated with variousconcentrations of plate-bound anti-CD3 Ab in the pres-ence of constant anti-CD28 (1/500 dilution of ascites)(A) or in the absence of anti-CD28 (B). Proliferationwas measured by [3H]thymidine incorporation duringthe final 18 h of a 72-h culture period. C and D, T cellsfrom OT-1 or OT-1 � c-FLIPL-Tg mice (5 � 104/well)were activated with either high-affinity OVAp (C) orlow-affinity variant SIIGFEKL (G4) (D) in the presenceof irradiated B6 spleen cells (3 � 105/well), and pro-liferation was measured by [3H]thymidine incorpora-tion during the final 18 h of a 72-h culture period. Thesefindings were consistent in five experiments.

FIGURE 2. Increased production and responsiveness to IL-2 byc-FLIPL-Tg CD8� T cells. A, Purified CD8� T cells (106/ml) were stim-ulated with the indicated concentrations of plate-bound anti-CD3 and aconstant amount of anti-CD28. After 48 h, supernatants were removed andanalyzed for IL-2 protein by ELISA. The findings were consistent in twoexperiments. B, CD8� T cells were activated with low-dose anti-CD3 (0.5�g/ml) in the presence of the indicated concentrations of rIL-2, and pro-liferation was measured after 3 days. C, T cells from OT-1 or OT-1 �c-FLIPL-Tg mice were activated with G4 peptide (10�7 M) in the presenceof the indicated concentrations of IL-2, and proliferation was measuredafter 3 days. Similar results were found in four additional experiments.

5272 c-FLIPL PROMOTES T CELL ACTIVATION VIA CASPASES

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

expressed over several days by T cells from c-FLIPL-Tg mice thanfrom control mice. By days 3 and 4 after activation, the percentageof CD25� cells became similar in the two groups, but this wasanticipated because during this period the CD25� cells would haveovergrown any cells lacking CD25. After CD25 levels began todecline on day 6 in control CD8� T cells, the enhanced expressionof CD25 became once again apparent in the c-FLIPL-Tg CD8� Tcells. However, during the entire period of analysis, the mean flu-orescence intensity of CD25 expression was higher on thec-FLIPL-Tg CD8� T cells (Fig. 3A, right panel).

The enhanced proliferative capacity of c-FLIPL-Tg CD8� Tcells was further shown to reside within the CD25� fraction. Tcells from c-FLIPL-Tg mice or normal littermate controls wereactivated for 3 days and then stained for expression of surfaceCD25, and cell cycling was defined by uptake of propidium iodide.As shown in Fig. 4, A and B, c-FLIPL-Tg T cells demonstrated anincreased proportion of CD25� cells after activation, and cell cy-

cling was confined to the CD25� subset. Furthermore, within theCD25� fraction of activated T cells, there was little difference incell cycle rates between c-FLIPL-Tg vs littermate control mice(Fig. 4C). Thus, the increased proliferation of c-FLIPL-Tg T cellswas also due in part to increased expression of CD25, and ratherthan IL-2-independent proliferation. This was also consistent withthe inhibition of proliferation of c-FLIPL-Tg T cells by blockingCD25 engagement (Fig. 4, D and E). These findings support theview that the increased cell cycling of T cells from c-FLIPL-Tgmice is due to increased signal pathways that lead to increasedCD25 expression, rather than effects directly on downstream cellcycle regulators that would be independent of CD25 expression.

In contrast to the increased production of IL-2 by c-FLIPL-TgCD8� T cells, levels of IFN-� were somewhat decreased, whereasIL-4 production was slightly increased (Fig. 5A). This was con-sistent at all doses of anti-CD3 tested (data not shown). This Tcytotoxic 2 pattern was confirmed by RNase protection analysis

FIGURE 3. Enhanced expressionof activation markers by stimulatedCD8� T cells from c-FLIPL-Tg mice.CD8� T cells from c-FLIPL-Tg micewere stimulated with anti-CD3 (5 �g/ml) and CD28 (1/500 dilution of as-cites), and surface expression of acti-vation molecules CD25 and CD69was measured by mean fluorescent in-tensity (MFI) by flow cytometry onthe days indicated. Similar resultswere obtained when results were ex-pressed as percent positive cells. Thestudies were performed three timeswith similar results.

FIGURE 4. Increased cell cycling of c-FLIPL-Tg Tcells is CD25-dependent. A—C, CD8� T cells wereactivated via anti-CD3 (5 �g/ml) and CD28 (1/500 di-lution of ascites) and analyzed on day 3 for DNA con-tent by propidium iodide (A), for surface expression ofCD25 (B), and DNA content gated on the CD25� sub-set (C). The gate for 2N DNA was established usingnaive noncycling T cells. D and E, Purified CD8� Tcells from normal littermate control (NLC) mice (D) orc-FLIPL-Tg mice (E) were activated via CD3/CD28 for3 days in the presence of the indicated concentrationsof control IgG or blocking anti-CD25 antibody. Theresults were consistent in three experiments.

5273The Journal of Immunology

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

(Fig. 5B). A similar pattern of cytokine production has been ob-served in the CD4� subset of c-FLIPL-Tg mice (18).

c-FLIPL-Tg CD8� T cells exhibit enhanced caspase and NF-�Bactivities that are independent of Fas

In considering the potential signaling pathways influenced by c-FLIPL in primary CD8� T cells, two aspects of c-FLIPL functionwere examined. First, we previously observed that transfection ofcell lines with c-FLIPL manifested increased ERK and NF-�B ac-tivities due to the association of c-FLIPL with, respectively, Raf-1and RIP1 (12). The second aspect was that c-FLIPL is known toassociate with and actually activate caspase-8 (23). c-FLIPL is alsoa potential substrate of caspase-8 (14–16). We therefore investi-gated to what extent these pathways might be related and contrib-ute to the phenotype of c-FLIPL-Tg T cells, and whether this re-quired the presence of Fas.

Consistent with the ability of c-FLIPL to activate caspase-8, c-FLIPL-Tg T cells manifested more caspase activity than littermatecontrol mice (Fig. 6A). This was apparent even in fresh resting Tcells, as well as in activated day 3 blasts. These findings weresupported by the observation that caspase-8 was more extensivelycleaved in c-FLIPL-Tg T cells, especially in the resting state (Fig.6B). Resting wild-type T cells contained largely full-lengthp55caspase-8, but following their activation, caspase-8 cleavage be-

came progressively apparent over the subsequent 3 days. By con-trast, c-FLIPL-Tg T cells demonstrated extensive cleavage ofcaspase-8 when freshly isolated, which did not increase as dramat-ically with activation as with wild-type T cells. This was likely dueto increased death of c-FLIPL-Tg T cells (A. Dohrman, unpub-lished observations), which would tend to constantly eliminatethose T cells with the highest levels of caspase activity. Becausedead cells were always intentionally eliminated from these analy-ses, it would remove this subset and further caspase-8 cleavagewould not be so apparent. Another issue is that caspase-8 can beactive in its full-length form when complexed to c-FLIPL (23), soanalysis of only the degree of caspase-8 cleavage may underesti-mate the amount of active caspase-8. The increased caspase activ-ity in c-FLIPL-Tg T cells was corroborated by increased and earlycleavage of c-FLIPL, a known caspase-8 substrate (Fig. 6B).

Neither the hyperproliferation nor the enhanced caspase activa-tion conferred by c-FLIPL required Fas. B6 c-FLIPL-Tg mice werecrossed with B6 lpr mice, and T cell proliferation was examined inresponse to varying doses of anti-CD3. As shown in Fig. 6C, in theabsence of Fas, purified CD8� T cells from c-FLIPL-Tg/lpr miceproliferated more extensively than those from age- and sex-matched littermate control lpr mice. This was similar to that ob-served earlier with c-FLIPL-Tg and wild-type mice. In parallelwith these findings, c-FLIPL-Tg/lpr CD8� T cells contained morecaspase activity than lpr T cells (Fig. 6D).

The presence of caspase activity was actually required for theproliferative response of c-FLIPL-Tg CD8� T cells, because thecaspase blockers z-Val-Ala-Asp(OCH3)-fluoromethylketone (z-VAD-fmk) and QVD-OPh largely inhibited both proliferation(Fig. 7A) and CD25 expression (B) in a dose-dependent manner,both in wild-type and c-FLIPL-Tg CD8� T cells. Similar resultswere observed for the CD4� subset (data not shown). These agentswere not merely toxic to lymphocytes, because there was no in-crease in dead cells in cultures containing caspase blockers com-pared with unstimulated T cells, and the delayed addition of z-VAD by even 24 h after activation resulted in substantially lessinhibition of T cell growth (data not shown). Caspase activity istherefore required particularly during the initial 24 h of T cellactivation. Furthermore, the augmented caspase activity ofc-FLIPL-Tg T cells is at least partly responsible for their increasedproliferative capacity.

FIGURE 5. T cytotoxic 2 pattern of cytokine production by c-FLIPL-TgCD8� T cells. Purified CD8� T cells (106/ml) were stimulated with anti-CD3 (5 �g/ml) plus anti-CD28 (1/500 dilution of ascites)/CD28, and su-pernatants were examined on days 2 and 3 for production of IL-4, IL-2, andIFN-�, by ELISA of culture supernatants (A), and RNase protection assayof RNA from the same cells (B). Similar results were observed in twoadditional experiments.

FIGURE 6. Caspase activity of CD8� T cells fromc-FLIPL-Tg mice is increased and independent of Fasexpression. A, Caspase activity was measured by rho-damine release from DEVD in freshly isolated CD8�

cells, and day 3 blasts were activated with anti-CD3 (5�g/ml) and CD28 (1/500 dilution of ascites) from nor-mal littermate controls (NLC) and c-FLIPL-Tg mice. B,Caspase activation as measured by Western blot show-ing cleavage of caspase-8 and c-FLIPL. The levels ofc-FLIPL in T cells from transgenic mice was 8- to 10-fold increased over normal littermate controls by den-sitometry. C and D, Absence of Fas in lpr mice does notreverse the hyperproliferation (C) or increased caspaseactivity (D) of c-FLIPL-Tg CD8� T cells. PurifiedCD8� cells were stimulated with the indicated concentra-tions of anti-CD3 (C) or 5 �g/ml anti-CD3 (D) plus anti-CD28 (1/500 dilution of ascites) as in Fig. 1A. These re-sults were similar in four additional experiments.

5274 c-FLIPL PROMOTES T CELL ACTIVATION VIA CASPASES

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Because the signal pathways that involve regulation of CD25and cytokines include the transcription factors NF-�B, AP-1, andNFAT, we examined these pathways by crossing the c-FLIPL-Tgmice with luciferase reporter mice transgenic for the DNA bindingsite for each of these transcription factors (19–21). Maximal lu-ciferase activity in T cells from these mice is detectable 2–3 daysafter activation, the time required for resting T cells to becomemetabolically active, and closely parallels DNA binding by elec-tromobility shift assay (19–21, 25) (M. Rincon, unpublished ob-servations). Luciferase activity was measured following activationof purified CD8� T cells with anti-CD3/CD28. An increase inNF-�B activity was observed on days 2 and 3 in CD8� T cellsfrom the NF-�B � c-FLIPL-Tg mice, and was statistically signif-icant (Fig. 8A). This was not merely due to the greater ability ofc-FLIPL-Tg T cells to produce IL-2, because the amount of anti-CD3 (5 �g/ml) was a dose at which no difference in IL-2 produc-tion was observed (see Fig. 5). Furthermore, addition of saturatinglevels of exogenous IL-2 yielded persistent increases in NF-�Bactivity in c-FLIPL-Tg CD8� T cells. By contrast, there were nosignificant differences in AP-1 or NFAT activity (Fig. 8A). Thesefindings were supported by observations of greater phosphoryla-tion and decreased levels of the NF-�B inhibitor, I�B�, inc-FLIPL-Tg T cells compared with wild-type T cells at all timepoints examined (Fig. 8B). Examination of more proximal signalsin the TCR pathway revealed no difference in either CD3� phos-phorylation or Ca2� flux following TCR stimulation (data notshown). These studies thus focused attention on the possible linkbetween c-FLIPL and the NF-�B pathway.

The requirement of caspase activity for T cell proliferationraised the possibility of a caspase substrate whose cleavage wouldpromote NF-�B activation. We have previously observed that, al-though transient transfection with both c-FLIPL and p43FLIP N-terminal fragment induces NF-�B activity, caspase inhibition byz-VAD blocked NF-�B activity by c-FLIPL but not by p43FLIP

(26). We have also observed in transient transfection studies thatc-FLIP associates with RIP1, a known activator of NF-�B (12).We therefore compared the ability of c-FLIPL vs p43FLIP to recruitRIP1. 293 T cells were transfected with c-FLIPL or p43FLIP. Co-immunoprecipitation studies demonstrated that endogenous RIP1associated with p43FLIP but not with full-length c-FLIPL (Fig. 9A).Furthermore, the association of p43FLIP with RIP1 required thepresence of caspase-8 because this association did not occur in acaspase-8-low variant of 293 cells, but was restored with cotrans-fection of caspase-8 (Fig. 9B). Moreover, a dominant-negativeform of RIP1 (RIP1559–671) was able to inhibit p43FLIP-inducedNF-�B activity, attesting to the view that this function of c-FLIPwas indeed working via RIP1 (Fig. 9C).

To further determine whether c-FLIPL and RIP1 were indeedassociated with active caspases in CD8� in T cells, fresh day 0 andday 4 c-FLIPL-Tg T cell blasts were incubated with biotin-VADfollowed by lysis and precipitation of active caspases with avidin-Sepharose. Similar to the findings using DEVD-rhodamine, therewas little active caspase-8 precipitated from day 0 T cells com-pared with day 4 blasts even though the amount of total caspase-8was identical in the two populations (Fig. 9D). The active form ofcaspase-8 in day 4 T cell blasts was full-length caspase-8, which is

FIGURE 7. Caspase-dependent proliferation andCD25 expression by T cells from c-FLIPL-Tg mice. Pu-rified CD8� T cells from normal littermate controls(NLC) or c-FLIPL-Tg mice were stimulated with 5�g/ml anti-CD3 plus anti-CD28 (1/500 dilution of as-cites) in the presence of the pancaspase blockersz-VAD-fmk or QVD-OPh, or controls DMSO or zFA.A, Proliferation measured by [3H]thymidine incorpora-tion during the final 18 h of a 72 h culture period. B,Surface CD25 expression on day 3. These findings wereconsistent in two experiments.

FIGURE 8. Increased NF-�B activity by c-FLIPL-TgCD8� T cells. A, c-FLIPL-Tg mice were crossed by micetransgenic for the DNA binding sites of AP-1, NF-�B, andNFAT linked to a luciferase reporter. Purified CD8� T cellswere activated with anti-CD3 (5 �g/ml) and CD28 (1/500dilution of ascites) and measured on days 1, 2, and 3 forluciferase activity. Results represent mean � SD of threeseparate experiments. �, Statistically significant differences(p � 0.05). B, CD8� T cells were used either freshly iso-lated or following activation with the same concentrationsof anti-CD3/CD28 as in A for 2 or 3 days, and cell lysateswere analyzed by Western blot for expression of phospho-I�B� (p-I�B�), total I�B�, and actin. The findings wereconsistent in three studies.

5275The Journal of Immunology

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

consistent with the modeling of caspase-8 with c-FLIPL (23). Inaddition, very little c-FLIP was associated with active caspase-8 inday 0 lysates, but increased considerably in day 4 lysates (Fig. 9D).Of interest was that the c-FLIP associated with caspase-8 wasmore the cleaved p43FLIP form than full-length c-FLIPL, whereasthese two forms were about equally present in the whole-cell ly-sates. This suggests that c-FLIPL that is in close proximity withactive caspase-8 is more likely to be cleaved. In addition, RIP1was also found associated with active caspase-8 in day 4 lysatesbut not in day 0 lysates. Finally, to examine the effect of caspaseblockage on activation of NF-�B in primary T cells, we used micetransgenic for the DNA binding site for NF-�B linked to the lu-ciferase gene (21). Purified CD8� T cells from these mice wereactivated with anti-CD3/CD28 for 48 h in the absence or presenceof the caspase blockers, z-VAD-fmk and QVD-OPh. As shown inFig. 9E, caspase inhibition greatly diminished the amount ofNF-�B activity observed in activated primary T cells. This is con-sistent with the requirement of caspase-dependent cleavage of c-FLIPL to recruit RIP1, and begins to provide an explanation for thecaspase requirement for T cell activation.

Discussionc-FLIPL was originally identified as an inhibitor of cell death in-duced by Fas through its ability to compete with caspase-8 forrecruitment to FADD during Fas ligation (10). More recently, twoadditional functions of c-FLIPL have been discerned. The first isthe association of c-FLIPL with Raf-1, thereby promoting activa-tion of ERK, as well as its association with TRAF2 and RIP1,

promoting activation of NF-�B (12). The second is the ability ofc-FLIPL to heterodimerize with and activate caspase-8 (23). Al-though the capacity of c-FLIPL to both compete with caspase-8 forrecruitment to FADD, as well as activate caspase-8 by direct het-erodimerization, may seem contradictory, the current findingsnonetheless suggest that at least one caspase-dependent activationpathway involves c-FLIPL as both an activator and substrate ofcaspase-8. Following caspase-dependent cleavage of c-FLIPL,RIP1 is recruited more efficiently to p43FLIP, thereby allowingmore effective activation of NF-�B.

The current findings extend in several ways our previous obser-vations that c-FLIPL can promote T cell growth. First, it demon-strates that caspase activity is increased in c-FLIPL-Tg T cells andthis is required to promote proliferation of c-FLIPL-Tg T cells,functioning largely through increased expression of CD25. Sec-ond, this is the first demonstration that activated T cells containactive caspase-8 in a full-length form and that c-FLIP and RIP1 areassociated with active caspases in cycling T cells, but not in naiveT cells, despite similar levels of these proteins.

Fas has been demonstrated to stimulate a variety of effectorfunctions in several cell types, including primary T cell prolifer-ation (5), fibroblast growth (3), hepatocyte regeneration (1), neu-rite outgrowth (2), and up-regulation of costimulatory moleculesand cytokines by dendritic cells (27). The physiological signifi-cance of these phenomena is uncertain, because Fas-deficient lprmice do not manifest defects in T cell proliferation, nor do theyhave overt developmental abnormalities in other organs systemsbeen reported in lpr mice. The signal pathway(s) responsible for

FIGURE 9. Requirement of caspase-8 to promote association of RIP1 with p43FLIP and activation of NF-�B. A and B, 293T cells (A) or caspase-8-lowvariant 293 cells (B, left panel) were transfected with FLAG-tagged caspase-8, p43caspase-8, c-FLIPL, or p43FLIP for 16 h, and cell lysates were precipitatedusing anti-FLAG followed by Western blot detection of RIP1. B, Right panel, Transfection of caspase-8 into caspase-8-low variant 293 cells restoresassociation of RIP1 with p43FLIP. Coimmunoprecipitation of p43FLIP was performed as in A and B in the absence (�) or presence (�) of transfectedcaspase-8. C, Dominant-negative RIP1559–671 was transfected using the amounts indicated into 293T cells along with p43FLIP followed by measurementof NF-�B-luciferase activity. D, Purified c-FLIPL-Tg T cells, either freshly isolated (D0) or activated with anti-CD3/CD28 plus IL-2 for 4 days (D4), wereincubated for 15 min with biotin-VAD-fmk (10 �M) and then lysed. Lysates were subjected to a preclear with Sepharose beads and then incubated withavidin-Sepharose. Precipitates were analyzed by immunoblot for caspase-8, c-FLIP, and RIP1. Shown are avidin-Sepharose precipitates compared withwhole-cell lysates (WCL) as a reference. E, CD8� T cells from NF-�B-luciferase mice were activated with anti-CD3 (5 �g/ml) and CD28 (1/500 dilutionof ascites) for 48 h in the absence or presence of caspase blockers z-VAD-fmk or QVD-OPh (each at 100 �M) and then assayed for luciferase activity.The results were similar in two additional studies.

5276 c-FLIPL PROMOTES T CELL ACTIVATION VIA CASPASES

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

this unanticipated stimulatory property of Fas is not well de-scribed. However, in the case of T cells and neurons, a link of Fasstimulation to ERK and NF-�B activation has been shown (2, 12).A more compelling case has been made for the involvement ofcaspase activity as a requirement for T cell activation. Not only docaspase blockers inhibit proliferation of primary human T cells (8,9), but both murine and human T cells deficient in functionalcaspase-8 manifest a pronounced defect in proliferation (7, 22).Although numerous caspase-8 substrates may be important for fullT cell activation, the current findings indicate that c-FLIPL is likelyone of these substrates.

The finding that activated c-FLIPL-Tg CD8� T cells expresshigher levels of IL-2 and surface CD25, and that the increasedcycling of activated c-FLIPL-Tg T cells occurs within the CD25�

subset, supports the view that c-FLIPL-induced hyperproliferationof T cells occurs primarily via augmented IL-2 signaling and lesslikely through alterations of other cell cycle regulatory proteins.This is also consistent with the ability of c-FLIPL to augment ac-tivation of the ERK and NF-�B pathways, which are involved withinduction of IL-2 and CD25 expression (28–30).

At present, it is unclear how signals initiated by TCR ligationlink to those induced by c-FLIPL. A logical consideration might bethrough TCR up-regulation of surface FasL, which would thenbind Fas and recruit c-FLIPL. However, it is clear from the currentstudies that c-FLIPL also augments proliferation in Fas-deficientlpr mice. Preliminary studies identify a complex of activecaspase-8, c-FLIPL, and RIP1 in lpr T cell blasts (J. Russell, un-published observations). This does not exclude the potential in-volvement of other death receptors in the activation of caspases ineffector T cells. Future studies are underway to examine the linksbetween TCR signals and caspase activation.

RIP1 associates directly with a heterodimer of caspase-8/c-FLIP, although it is not certain to which component of the het-erodimer it binds. Given that the form of c-FLIP associated withactive caspase-8 is preferentially cleaved p43FLIP (Fig. 9D), com-bined with the greater association of RIP1 with transfected p43FLIP

rather than full-length c-FLIPL, this suggests that p43FLIP maystabilize binding of RIP1 to the caspase-8/c-FLIP heterodimer.This immediately suggests that c-FLIPL may be an important sub-strate for caspase-8 during T cell activation. An additional possi-bility is that the form of c-FLIP in the heterodimer may influencethe ability of HSP90 to stabilize RIP1 (31, 32). Thus, the amountof RIP1 complexed to caspase-8/c-FLIP may reflect differences indegradation more than association.

Other studies by us have shown a similar association of TRAF2preferentially to p43FLIP (26). c-FLIPL has a known caspase cleav-age site at Asp376, which yields p43FLIP (23). This exposes a bind-ing site for RIP1 and TRAF2 that is presumably less accessible infull-length c-FLIPL. These findings would also serve to explain theability of caspase inhibition to block NF-�B activation by c-FLIPL

but not by p43FLIP (26). Differences in caspase activity betweenCD8� and CD4� T cells might also explain why we observedincreased NF-�B activity in CD8� cells from c-FLIPL-Tg mice,but did not in previous studies of the CD4� subset fromc-FLIPL-Tg mice (18). In separate studies, we observe that acti-vated CD8� cells from c-FLIPL-Tg mice manifest greater caspaseactivity than activated CD4� cells from the same mice (A. Dohr-man, unpublished observations). In addition, p43FLIP lacks the do-main by which c-FLIPL activates caspase-8 (23). Thus, cleavage ofc-FLIPL to p43FLIP may serve not only to recruit RIP1 and TRAF2to promote NF-�B activation, it may also impede further activationof caspase-8, to decrease the risk of promoting T cell death afteractivation. In this regard, it is of some interest that the viral formof FLIP (v-FLIP) is truncated, containing only the two death do-

mains (33). v-FLIP still allows recruitment of RIP1 and TRAF2,similar to p43FLIP, but does not activate caspase-8. This couldprovide a survival capacity of v-FLIP.

Regulation of c-FLIPL not only in T cells but also in other celltypes may strongly influence the ability to either proliferate ordifferentiate. Blood monocytes have very low levels of c-FLIPL

and are highly sensitive to Fas-induced death, whereas dendriticcells that are derived from blood monocytes using GM-CSF andIL-4 express high levels of c-FLIPL and are highly resistant toFas-mediated apoptosis (Ref. 34; R. Budd, unpublished observa-tions). Furthermore, ligation of Fas on dendritic cells promotesup-regulation of surface CD80/CD86, class II MHC, and IL-12production (27). A murine model of cardiac hypertrophy has alsoimplicated a role for Fas in cardiac hypertrophy in response tohypertension (4). Interestingly, cardiac myocytes are among thehighest expressors of c-FLIPL, and FLIP-null mice are embryoniclethal due to a cardiac developmental abnormality (35). Thus, c-FLIP is emerging as not only an inhibitor of Fas-induced death, butalso as a promoter of cell proliferation and/or differentiation basedon its ability to recruit adaptor proteins linking to the ERK andNF-�B pathways. That this occurs better with cleaved p43FLIP thanwith full-length c-FLIPL suggests that c-FLIPL is at least onecaspase-8 substrate needed for optimal T cell proliferation.

AcknowledgmentsWe thank Colette Charland for technical assistance with flow cytometry.

DisclosuresThe authors have no financial conflict of interest.

References1. Desbarats, J., and M. K. Newell. 2000. Fas engagement accelerates liver regen-

eration after partial hepatectomy. Nat. Med. 6:920.2. Desbarats, J., R. B. Birge, M. Mimouni-Rongy, D. E. Weinstein, J. S. Palerme,

and M. K. Newell. 2003. Fas engagement induces neurite growth through ERKactivation and p35 upregulation. Nat. Cell Biol. 5:118.

3. Aggarwal, B. B., S. Singh, R. LaPushin, and K. Totpal. 1995. Fas antigen signalsproliferation of normal human diploid fibroblast and its mechanism is differentfrom tumor necrosis factor receptor. FEBS Lett. 364:5.

4. Badorff, C., H. Ruetten, S. Mueller, M. Stahmer, D. Gehring, F. Jung, C. Ihling,A. M. Zeiher, and S. Dimmeler. 2002. Fas receptor signaling inhibits glycogensynthase kinase 3� and induces cardiac hypertrophy following pressure overload.J. Clin. Invest. 109:373.

5. Alderson, M. R., R. J. Armitage, E. Maraskovsky, T. W. Tough, E. Roux,K. Schooley, F. Ramsdell, and D. H. Lynch. 1993. Fas transduces activationsignals in normal human T lymphocytes. J. Exp. Med. 178:2231.

6. Yeh, W. C., J. L. Pompa, M. E. McCurrach, H. B. Shu, A. J. Elia, A. Shahinian,M. Ng, A. Wakeham, W. Khoo, K. Mitchell, et al. 1998. FADD: essential forembryo development and signaling from some, but not all, inducers of apoptosis.Science 279:1954.

7. Chun, H. J., L. Zheng, M. Ahmad, J. Wang, C. K. Speirs, R. M. Siegel, J. K. Dale,J. Puck, J. Davis, C. G. Hall, et al. 2002. Pleiotropic defects in lymphocyteactivation caused by caspase-8 mutations lead to human immunodeficiency. Na-ture 419:395.

8. Kennedy, N. J., T. Kataoka, J. Tschopp, and R. C. Budd. 1999. Caspase activationis required for T cell proliferation. J. Exp. Med. 190:1891.

9. Alam, A., L. Y. Cohen, S. Aouad, and R. P. Sekaly. 1999. Early activation ofcaspases during T lymphocyte stimulation results in selective substrate cleavagein nonapoptotic cells. J. Exp. Med. 190:1879.

10. Irmler, M., M. Thome, M. Hahne, P. Schneider, K. Hofmann, V. Steiner, J.-L.Bodmer, M. Schroter, K. Burns, C. Mattmann, et al. 1997. Inhibition of deathreceptor signals by cellular FLIP. Nature 388:190.

11. Tschopp, J., M. Irmler, and M. Thome. 1998. Inhibition of fas death signals byFLIPs. Curr. Opin. Immunol. 10:552.

12. Kataoka, T., R. C. Budd, N. Holler, M. Thome, F. Martinon, M. Irmler, K. Burns,M. Hahne, N. Kennedy, M. Kovacsovics, and J. Tschopp. 2000. The caspase-8inhibitor FLIP promotes activation of NF-�B and Erk signaling pathways. Curr.Biol. 10:640.

13. Lens, S. M. A., T. Kataooka, K. A. Fortner, A. Tinel, I. Ferrero,H. R. MacDonald, M. Hahne, F. Beermann, A. Attinger, H.-A. Orbea, et al. 2002.The caspase-8 inhibitor c-FLIPL modulates T-cell receptor-induced proliferationbut not activation-induced cell death of lymphocytes. Mol. Cell. Biol. 22:5419.

14. Chang, D. W., Z. Xing, Y. Pan, A. Algeciras-Schimnich, B. C. Barnhart,S. Yaish-Ohad, M. E. Peter, and X. Yang. 2002. c-FLIPL is a dual functionregulator for caspase-8 activation and CD95-mediated apoptosis. EMBO J. 21:3704.

5277The Journal of Immunology

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

15. Krueger, A., I. Schmitz, S. Baumann, P. H. Krammer, and S. Kirchhoff. 2001.Cellular FLICE-inhibitory protein splice variants inhibit different steps ofcaspase-8 activation at the CD95 death-inducing signaling complex. J. Biol.Chem. 276:20633.

16. Scaffidi, C., I. Schmitz, P. H. Krammer, and M. E. Peter. 1999. The role of c-FLIPin modulation of CD95-induced apoptosis. J. Biol. Chem. 274:1541.

17. Hogquist, K. A., S. C. Jameson, W. R. Heatyh, J. L. Howard, M. J. Bevan, andF. R. Carbone. 1994. T cell receptor antagonist peptides induce positive selection.Cell 76:17.

18. Wu, W., L. Rinaldi, K. A. Fortner, J. Q. Russell, J. Tschopp, C. Irvin, andR. C. Budd. 2004. Cellular FLIP long form-transgenic mice manifest a Th2 cy-tokine bias and enhanced allergic airway inflammation. J. Immunol. 172:4724.

19. Rincon, M., B. Derijard, C. W. Chow, R. J. Davis, and R. A. Flavell. 1997.Reprogramming the signalling requirement for AP-1 (activator protein-1) acti-vation during differentiation of precursor CD4� T-cells into effector Th1 and Th2cells. Genes Funct. 1:51.

20. Rincon, M., and R. A. Flavell. 1997. Transcription mediated by NFAT is highlyinducible in effector CD4� T helper 2 (Th2) cells but not in Th1 cells. Mol. Cell.Biol. 17:1522.

21. Voll, R. E., E. Jimi, R. J. Phillips, D. F. Barber, M. Rincon, A. C. Hayday,R. A. Flavell, and S. Ghosh. 2000. NF-�B activation by the pre-T cell receptorserves as a selective survival signal in T lymphocyte development. Immunity13:677.

22. Salmena, L., B. Lemmers, A. Hakem, E. Matysiak-Zablocki, K. Murakami,P. Y. Au, D. M. Berry, L. Tamblyn, A. Shehabeldin, E. Migon, et al. 2003.Essential role for caspase 8 in T-cell homeostasis and T-cell-mediated immunity.Genes Dev. 17:883.

23. Micheau, O., M. Thome, P. Schneider, N. Holler, J. Tschopp, D. W. Nicholson,C. Briand, and M. G. Grutter. 2002. The long form of FLIP is an activator ofcaspase-8 at the Fas death-inducing signaling complex. J. Biol. Chem.277:45162.

24. Alam, S. M., G. M. Davies, C. M. Lin, T. Zal, W. Nasholds, S. C. Jameson,K. A. Hogquist, N. R. Gascoigne, and P. J. Travers. 1999. Qualitative and quan-titative differences in T cell receptor binding of agonist and antagonist ligands.Immunity 10:227.

25. Auphan, N., A. K. Simon, H. Asnagli, R. J. Phillips, M. Rincon, S. Ghosh,R. A. Flavell, and A. M. Schmitt-Verhulst. 1998. Consequences of intrathymic

TCR engagement by partial agonist on selection events and peripheral T cellactivation program. J. Immunol. 160:4810.

26. Kataoka, T., and J. Tschopp. 2004. N-Terminal fragment of c-FLIPL processedby caspase 8 specifically interacts with TRAF2 and induces activation of theNF-�B signaling pathway. Mol. Cell. Biol. 24:2627.

27. Rescigno, M., V. Piguet, B. Valzasina, S. Lens, R. Zubler, L. French, V. Kindler,J. Tschopp, and P. Ricciardi-Castagnoli. 2000. Fas engagement induces the mat-uration of dendritic cells (DCs), the release of interleukin (IL)-1�, and the pro-duction of interferon-� in the absence of IL-12 during DC-T cell cognate inter-action: a new role for fas ligand in inflammatory responses. J. Exp. Med. 192:1661.

28. Cross, S. L., N. F. Halden, M. J. Lenardo, and W. J. Leonard. 1989. Functionallydistinct NF-�B binding sites in the immunoglobulin � and IL-2 receptor � chaingenes. Science 244:466.

29. Li, Y. Q., C. S. Hii, C. J. Der, and A. Ferrante. 1999. Direct evidence that ERKregulates the production/secretion of interleukin-2 in PHA/PMA-stimulated Tlymphocytes. Immunology 96:524.

30. Hughes, C. C., and J. S. Pober. 1996. Transcriptional regulation of the interleu-kin-2 gene in normal human peripheral blood T cells: convergence of costimu-latory signals and differences from transformed T cells. J. Biol. Chem. 271:5369.

31. Lewis, J., A. Devin, A. Miller, Y. Lin, Y. Rodriguez, L. Neckers, and Z. G. Liu.2000. Disruption of hsp90 function results in degradation of the death domainkinase, receptor-interacting protein (RIP), and blockage of tumor necrosis factor-induced nuclear factor-�B activation. J. Biol. Chem. 275:10519.

32. Zhao, C., and E. Wang. 2004. Heat shock protein 90 suppresses tumor necrosisfactor-�-induced apoptosis by preventing the cleavage of Bid in NIH3T3 fibro-blasts. Cell. Signal. 16:313.

33. Thome, M., P. Schneider, K. Hofmann, H. Fickenscher, E. Meinl, F. Neipel,C. Mattmann, K. Burns, J.-L. Bodmer, M. Schroter, et al. 1997. Viral FLICE-inhibitory proteins (FLIPs) prevent apoptosis induced by death receptors. Nature386:517.

34. Ashany, D., A. Savir, N. Bhardwaj, and K. B. Elkon. 1999. Dendritic cells areresistant to apoptosis through the Fas (CD95/APO-1) pathway. J. Immunol. 163:5303.

35. Yeh, W. C., A. Itie, A. J. Elia, M. Ng, H. B. Shu, A. Wakeham, C. Mirtsos,N. Suzuki, M. Bonnard, D. V. Goeddel, and T. W. Mak. 2000. Requirement forCasper (c-FLIP) in regulation of death receptor-induced apoptosis and embryonicdevelopment. Immunity 12:633.

5278 c-FLIPL PROMOTES T CELL ACTIVATION VIA CASPASES

by guest on April 4, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from