A Costimulatory Function for CD40 Melissa E. Munroe and Gail A. Bishop J Immunol 2007; 178:671-682; ; This information is current as doi: 10.4049/jimmunol.178.2.671 of September 29, 2021. http://www.jimmunol.org/content/178/2/671 Downloaded from References This article cites 96 articles, 50 of which you can access for free at: http://www.jimmunol.org/content/178/2/671.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/ • 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

*average by guest on September 29, 2021 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 © 2007 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

A Costimulatory Function for T Cell CD401

Melissa E. Munroe* and Gail A. Bishop2*†‡

CD40 plays a significant role in the pathogenesis of inflammation and autoimmunity. CD40 directly activates cells, which can result in autoantibody production. T cells can also express CD40, with an increased frequency and amount of expression seen -in CD4؉ T of autoimmune mice, including T cells from mice with collagen-induced arthritis. However, the mech anisms of T cell CD40 function have not been clearly defined. To test the hypothesis that CD40 can serve as a costimulatory molecule on T lymphocytes, CD40؉ T cells from collagen-induced arthritis mice were examined in parallel with mouse and human T cell lines transfected with CD40. CD40 served as effectively as CD28 in costimulating TCR-mediated activation, including induction of and transcription factor activities and production of . An additional enhancement was seen when both CD40 and CD28 signals were combined with AgR stimulation. These findings reveal potent biologic functions for T cell CD40 and suggest an additional means for amplification of autoimmune responses. The Journal of Immunology, 2007, 178: 671–682. Downloaded from he 50-kDa membrane of the TNFR superfamily, family members as costimulatory molecules for Ag receptor stim- CD40, is expressed by APC, including dendritic cells, ulation (reviewed in Refs. 21 and 22). We hypothesized that co- T , and B lymphocytes (1, 2). The ligand for stimulation is a plausible role for CD40 on T cells. In the studies CD40, CD154, is expressed on activated T cells and allows for presented here, we determined that T cell CD40 augmented CD3 interactions with APC during the cognitive phase of the immune and CD3 plus CD28-mediated production in T cells from 3

response, as well as directing effector T cell-dependent B cell ac- mice which have developed collagen-induced arthritis (CIA), as http://www.jimmunol.org/ tivation (3). Such interactions have been directly implicated in well as in T cell lines stably transfected with CD40. Although autoimmunity. Blocking CD40-CD154 interactions has been CD40 signals alone did not activate NFAT or IL-2 secretion, CD40 shown to either prevent or alleviate such diseases as insulin de- ligation markedly augmented CD3 and CD3 plus CD28 responses. pendent diabetes mellitus (4), arthritis (5), and systemic lupus er- As in B cells, T cell CD40 was able to efficiently bind the adaptor ythematosus (6), the latter two benefiting from abrogated T cell- TNFR-associated factors (TRAF), activate NF-␬B and dependent autoantibody production (7, 8). AP-1 pathways, and stimulate TNF-␣ secretion. Taken together, The role of CD40 as a direct signal receptor has now been ex- these findings reveal that CD40 can act as a powerful signaling panded to T cells. Shortly after CD154 was cloned, it was dem- receptor on T as well as B lymphocytes, a function that may have

onstrated that CD154 can augment mitogen and TCR-mediated important implications for T-B interactions in autoimmune by guest on September 29, 2021 proliferation of CD4ϩ and CD8ϩ T cells (9), although the lack of diseases. CD154-specific Abs at that time precluded further investigation. Although a specific biologic role for CD40 on CD8ϩ T cells re- Materials and Methods mains undefined (10–13), it has been shown that autoimmune- Cells ϩ ϩ prone strains of mice have increased numbers of CD40 CD4 T The mouse T cell line 2B4.11 (23) and human T cell line Jurkat (24) have cells compared with normal strains (14). The most extensively been described previously. Cell lines and their stable transfectants express- studied of these is the NOD mouse (15–17). ing hCD40 were maintained in RPMI 1640 containing 10% FCS (Hy- ␮ To date, the physiologic role(s) of CD40 on T cells has not been Clone), 10 M 2-ME, and antibiotics. These subclones are referred to as 2B4.hCD40 and J.hCD40. Hi5 insect cells expressing hCD154 have been characterized, nor have the mechanisms by which CD40 affects T described and characterized previously (25, 26). These cells grow at 26°C cell function been defined. We have previously studied CD40 as an and rapidly die to form membrane fragments at 37°C and therefore do not important signaling molecule on B lymphocytes, delivering signals overgrow cell cultures. ␬ alone and synergistically with the BCR (18), leading to NF- B and Stable transfections JNK pathway activation and subsequent proliferation, secretion of cytokines, Ig production and isotype switching (reviewed in Refs. Cell lines were stably transfected with a previously reported hCD40 ex- pression plasmid (27) as described previously (28). G418-resistant clones 19 and 20). Like B cells, T cells have been shown to use TNFR were analyzed for expression of hCD40 using a FACScan flow cytometer (BD Biosciences) and mean channel fluorescence (MCF) determined using FlowJo software. *Department of Microbiology and †Department of Internal Medicine, University of Reagents Iowa, Iowa City, IA 52242; and ‡Veterans Affairs Medical Center, Iowa City, IA 52242 Recombinant mouse TNF-␣, IL-2, and IFN-␥ were purchased from Pep- Received for publication June 21, 2006. Accepted for publication October 17, 2006. roTech. Streptavidin-HRP was purchased from Jackson ImmunoResearch Laboratories. ELISA TMB peroxidase substrate was purchased from KPL. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance Tosylactivated Dynabeads for Ab conjugation (per manufacturer’s instruc- with 18 U.S.C. Section 1734 solely to indicate this fact. tions) were purchased from Dynal Biotech. PMA and ionomycin were purchased from Sigma-Aldrich. 1 This work was supported by grants from the National Institutes of Health and the Veterans’ Administration (to G.A.B.) and postdoctoral fellowship support provided by the American Heart Association and the American Cancer Society (to M.E.M.). 3 Abbreviations used in this paper: CIA, collagen-induced arthritis; CII, type II 2 Address correspondence and reprint requests to Dr. Gail A. Bishop, 2193B MERF, chicken collagen; MCF, mean channel fluorescence; TRAF, TNFR-associated factor. Department of Microbiology, University of Iowa, Iowa City, IA 52242. E-mail ad- dress: [email protected] Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00

www.jimmunol.org 672 SIGNALING BY CD40 IN T CELLS

Antibodies T cell isolation The 1C10 (anti-mCD40, rat IgG2a), 72-2 (rat IgG2a isotype control), and T cells were isolated from mouse spleens 70 days postimmunization. G28-5 (anti-hCD40, mouse IgG1) hybridomas were purchased from the Briefly, spleens from euthanized mice were teased apart with forceps, American Type Culture Collection. MOPC-31c (mouse IgG1 isotype con- erythrocytes lysed in ACK buffer, and remaining cells placed over a T cell trol) was from Sigma-Aldrich. Polyclonal rabbit anti-TRAF2 Ab was from enrichment column per manufacturer’s protocol (R&D Biosystems). T MBL. Polyclonal mouse anti-yy1 and polyclonal rabbit anti-TRAF3, anti- cells were enriched to ϳ90% purity as determined by flow cytometry (see TRAF1, anti-TRAF6, and anti-hCD40 Abs were from Santa Cruz Biotech- Table I). Primary mouse T cells were cultured in Click’s medium contain- nology. Polyclonal rabbit anti-I␬B␣, anti-phosphorylated I␬B␣, anti- ing 1% nutridoma-SP (Roche), 10 ␮M 2-ME, and antibiotics, or stained for NF␬B2 p100/p52, anti-JNK, and anti-phosphorylated JNK Abs were from CD3␧ (PE) and CD40 (FITC) and analyzed by flow cytometry. Technology. Mouse anti-actin Ab (C4) was from Chemicon International. Peroxidase-labeled goat anti-rabbit and goat anti-mouse IgG Abs were from Jackson ImmunoResearch Laboratories. Anti-mouse CD3␧ NF-␬B/NFAT/AP-1 dual luciferase reporter assays (145-2C11; Armenian hamster IgG1), anti-human CD3 (OKT3; mouse IgG2a), anti-mouse CD28 (37.51; hamster IgG), anti-human CD28 2B4.hCD40 or J.hCD40 cells (1.5 ϫ 107) were transiently transfected with (CD28.2; mouse IgG1), and relevant isotype control Abs were purchased 20 ␮gof4ϫ NF-␬B, 40 ␮gof4ϫ NFAT, or 40 ␮gof7ϫ AP-1 luciferase from eBioscience. PE-labeled anti-mouse CD3 (145-2C11; Armenian ham- reporter plasmid (31), and 1 ␮gofRenilla luciferase vector (pRL-null; ster IgG1), FITC labeled anti-mouse CD40 (HM40-3; Armenian hamster Promega) by electroporation. Cells were rested on ice for 15 min, then IgM), anti-human CD40 (5C3; mouse IgG3), anti-human CD154 (TRAP1, stimulated (2 ϫ 106 cells/ml) for 6 h (NF-␬B) or 24 h (NFAT/AP-1) with mouse IgG1), anti-mouse CD80 (16-10A1; Armenian hamster IgG2), anti- 10 ␮g/ml anti-hCD40 or isotype controls, and/or 5 ϫ 105 beads/ml anti- mouse CD86 (GL1; rat IgG2a), anti-mouse CD95 (Jo2; Armenian hamster CD3 or anti-CD3ϩCD28-coated Dynabeads. After stimulation, cells were IgG2), and relevant isotype control Abs were purchased from BD Pharm- pelleted, lysed, and assayed for relative luciferase activity (NF-␬B, NFAT,

ingen. FITC labeled anti-mouse CD25 (PC61.5; rat IgG1), anti-mouse or AP-1:Renilla) per manufacturer’s protocol (Promega) using a Turner Downloaded from CD54 (YN1/1.7.4; rat IgG2b), and anti-mouse CD11␣ (M17/4; rat IgG2a) Designs 20/20 luminometer, with settings of a 2-s delay followed by a Abs were purchased from eBioscience. Biotin-labeled anti-mouse CD154 10-s read. (MR1; Armenian hamster IgG) and relevant isotype control Abs were pur- chased from eBioscience. Alexa Fluor 488-labeled streptavidin was pur- chased from Molecular Probes/Invitrogen Life Technologies. Anti-mouse I␬B␣/JNK assays IL-2 and IFN-␥ (coating and biotinylated) ELISA Abs were purchased from Caltag Laboratories. Anti-human IL-2 and anti-mouse TNF-␣ (coat- 2B4.hCD40 or J.hCD40 cells (2 ϫ 106) were stimulated for indicated times ing and biotinylated) ELISA Abs were purchased from eBioscience. with culture medium, 10 ␮g of anti-hCD40 Ab (or respective isotype con- http://www.jimmunol.org/ trols) and/or 5 ϫ 105 beads/ml anti-CD3 or anti-CD3 plus CD28-coated Mice/CIA induction Dynabeads) to induce phosphorylation and degradation of the proteins blotted. The cells were pelleted by centrifugation, lysed and analyzed by Female C57BL/6 mice were purchased at 5–8 wk of age from the National SDS PAGE and Western blotting. Peroxidase-labeled Abs were visualized Cancer Institute. Mice were housed in a specific pathogen-free barrier fa- on Western blots using a chemiluminescent detection reagent (Pierce). cility with restricted access, and all procedures were performed as ap- proved by the University of Iowa Animal Care and Use Committee. CIA was induced based on the methods of Campbell et al. (29). Briefly, mice NF-␬B2 activation were either left naive, immunized in the tail s.c. with 100 ␮g of type II chicken collagen (CII; Sigma-Aldrich) dissolved in 10 mM acetic acid and 2B4.hCD40 or J.hCD40 cells (2 ϫ 106) were stimulated for indicated times emulsified in IFA (Sigma-Aldrich) containing 5 mg/ml H37 RA heat-killed with culture medium, 10 ␮g of anti-hCD40 Ab (or respective isotype con- by guest on September 29, 2021 mycobacteria (CFA; Difco Laboratories), or immunized with 10 mM acetic trols) and/or 5 ϫ 105 beads/ml anti-CD3 or anti-CD3 plus CD28-coated acid emulsified in CFA. Mice were monitored for limb erythema and swell- Dynabeads) to induce RelB activation, and processing of p100 to p52. The ing and paws measured (each paw recorded individually; four measure- cells were pelleted by centrifugation and cytoplasmic and nuclear fractions ments per mouse) with calipers two to three times per week (4026F; Mi- isolated as described previously (32). Samples were analyzed by SDS- tutoyo) (29, 30). All mouse studies were reviewed and approved by the PAGE and Western blotting. Peroxidase-labeled Abs were visualized on University of Iowa Animal Care and Use Committee. Western blots using a chemiluminescent detection reagent.

FIGURE 1. CD40 expressed on T cells from mice with CIA. A, C57BL/6 mice were immunized with CFA only, CII plus CFA, or remained naive, as described in Materials and Methods. Paws were measured two to three times per week from days 21–70 postimmuniza- tion. Mice receiving CII plus CFA had significant (p Ͻ 0.001) paw swelling compared with CFA or naive con- trols. Data represent two experiments (4 mice/group/ experiment; 8 mice/32 paws/data point total). B, Spleens from mice in A were pooled (2 spleens/pool, 4 pools/group), and T cells were isolated as described in Materials and Methods. Cells were stained with anti- mouse CD40 (FITC) and anti-mouse CD3␧ (PE) or iso- type control Abs and analyzed by flow cytometry. Quadrants were drawn based on staining with isotype control Abs. The Journal of Immunology 673

Table I. Increased number of CD40ϩ T cells in CIA micea

CII/CFA CFA Only Naive

Splenocytes (pre-T cell isolation) Percentage of CD3ϩ T cells 33.19 Ϯ 0.63 33.52 Ϯ 0.28 34.44 Ϯ 0.58 No. of CD3ϩ T cells (ϫ106) 5.67 Ϯ 0.17 5.11 Ϯ 0.17 4.81 Ϯ 0.35 Percentage of CD3/CD40ϩ T cells 9.31 Ϯ 0.52b 4.30 Ϯ 0.37 4.02 Ϯ 0.46 No. of CD3/CD40ϩ T cells (ϫ105) 15.90 Ϯ 1.54c 6.55 Ϯ 0.60 6.06 Ϯ 0.85 Negatively selected T cells Percentage of CD3ϩ T cells 88.67 Ϯ 1.56 87.41 Ϯ 3.28 89.01 Ϯ 1.67 No. of CD3ϩ T cells (ϫ106) 3.85 Ϯ 1.75 3.24 Ϯ 0.30 3.68 Ϯ 0.34 Percentage of CD3/CD40ϩ T cells 7.53 Ϯ 0.70d 2.96 Ϯ 0.26 2.80 Ϯ 0.56 No. of CD3/CD40ϩ T cells (ϫ105) 3.31 Ϯ 0.48e 1.11 Ϯ 0.17 1.40 Ϯ 0.11

a T cells were isolated as described in Materials and Methods from spleens of C57BL/6 mice receiving CII/CFA, CFA only, or remaining naive (four mice per group for two experiments; two spleens per pool, four pools total) 70 days postimmunization. Cells were stained with anti-mouse CD40 (FITC) and anti-mouse CD3⑀ (PE) or isotype control mAbs and analyzed by flow cytometry. Spleens from CII/CFA mice had significantly more CD3ϩCD40ϩ T cells than either CFA or naïve controls. b p Ͻ 0.01 (CII/CFA compared with CFA only or naive). c p Ͻ 0.001 (CII/CFA compared with CFA only or naive). d p ϭ 0.01 (CII/CFA compared with CFA only or naive). e p Ͻ 0.05 (CII/CFA compared with CFA only or naive). Downloaded from

Cytokine ELISA Results Primary mouse T cells or 2B4.hCD40 or J.hCD40 cells (4 ϫ 105) were CD40 expression on T cells from mice with CIA stimulated at optimal, empirically derived time points with culture me- CIA is a frequently used mouse model of inflammatory rheumatoid dium, 1 ␮g/ml anti-hCD40 Ab and/or anti-CD28, or plate-bound anti-CD3 (or respective isotype controls). Cytokine concentrations in culture super- arthritis that is both Ab and T cell dependent (35, 36). It has been http://www.jimmunol.org/ natants were determined by ELISA, using cytokine-specific coating Abs previously demonstrated that mouse strains prone to autoimmune and biotinylated detection Abs. Streptavidin-HRP binding to biotinylated diabetes have an increased number of T cells expressing CD40 that detection Abs was visualized with TMB substrate and the reaction was correlates with development of pathology (14). We examined stopped with 0.18 M H2SO4. Plates were read at 450 nm by a Spectra- Max250 Reader (Molecular Devices). Data were analyzed with SoftMax Pro software (Molecular Devices); unknowns were compared with a stan- dard curve containing at least five to seven dilution points of the relevant recombinant cytokine on each assay plate. In all cases, the coefficient of determination for the standard curve (r2) was Ͼ0.98. ELISA unknowns were diluted to fall within the standard values. by guest on September 29, 2021 TRAF recruitment to receptors in detergent-insoluble microdomains (Rafts) and immunoprecipitation. 2B4.hCD40 or J.hCD40 (1 ϫ 107) cells were stimulated with 10 ␮g of anti-hCD40 Ab (or isotype control Abs) or Hi5 cells ex- pressing hCD154 (or Hi5 cells expressing WT baculovirus; 1:4 Hi5 cells:lymphocytes) for 15 min at 37°C to induce recruitment of TRAFs to membrane rafts and allow formation of CD40 signaling complexes, as described previously (33). Detergent (1% Brij 58)- soluble and insoluble fractions were separated as described previ- ously (34). Samples of soluble and insoluble lysates were reserved for SDS-PAGE separation and analysis by Western blotting. The remainder of the lysates were immunoprecipitated with G-Sepharose beads (Amersham Biosciences) prewarmed with anti- hCD40 Ab for3hat4°C. The immunoprecipitation complexes were washed four times with lysis buffer before separation by SDS-PAGE and analysis by Western blot.

Up-regulation of cell surface proteins In experiments evaluating activation-induced up-regulation of surface pro- teins, 2B4.11 or 2B4.hCD40 cells were incubated with the indicated stimuli in 96-well plates (1–2 ϫ 105 cells/well). After 48–72 h, the cells were washed, then incubated for 20 min on ice in PBS-0.5% FCS-0.02% sodium FIGURE 2. azide containing 2.5 mM EDTA. EDTA treatment helped to dissociate cell CD40 enhances CD3 and CD3 plus CD28-mediated cyto- aggregates formed upon CD40 stimulation. Following the EDTA incuba- kine production in T cells from mice with CIA. Splenic T cells from tions, cells were washed and stained (in the absence of EDTA) with Abs for C57BL/6 mice immunized with CII plus CFA, CFA only, or remaining analysis by flow cytometry. naive were isolated as described in Materials and Methods. Cells were stimulated with plate-bound anti-CD3 Ϯ anti-CD28 and/or anti-mouse Statistical analyses CD40 Abs (compared with medium (Med)/isotype control (IC)). Culture ␥ Analyses were performed with GraphPad Instat software. A two-tailed supernatants were collected and assayed for IL-2 (A, 48 h), IFN- (B,72h), paired Student’s t test was used to determine significance between groups or TNF-␣ (C, 24 h) by ELISA. Data represent the mean Ϯ SEM of du- in CIA experiments, for cytokine ELISA, surface molecule up-regulation plicate samples of three independent experiments. There was no difference experiments, and luciferase reporter assays. between medium and isotype control samples. 674 SIGNALING BY CD40 IN T CELLS whether this was also true during the inflammatory process of CIA tively small percentage of T cells expressing CD40 (Fig. 1 and development. C57BL/6 mice injected with CII/CFA developed Table I). This small percentage limited detailed molecular charac- significant paw swelling ( p Ͻ 0.001) compared with mice given terization of CD40 function on CD40-expressing T cells, although CFA only or naive controls (Fig. 1A). Splenic T cells from these the data presented in Fig. 2 indicate that this population has sig- mice were isolated and evaluated for dual expression of CD3␧ and nificant biologic activity distinct from that of CD4ϩ T cells that do CD40. A representative FACS plot of isolated T cells is presented not express CD40. We thus wanted to complement these experi- in Fig. 1B; quantitation is presented in Table I. Small numbers of ments with stimulation of homogeneous populations of CD40ϩ T cells in the FACS samples that were CD3␧Ϫ were also CD40Ϫ and cells, as well as explore CD40 signaling pathways. Because of therefore unlikely to be APC (Fig. 1B and Table I, line 5). Data in limiting numbers of CD40ϩ T cells in CIA mice that would require Table I demonstrate that there are more than twice as many CD40ϩ potentially function-altering positive selection for isolation, T cells in the spleens of mice that received CII/CFA compared CD40ϩ T cell lines were a desirable alternative. We thus stably with mice that received CFA only or naive controls ( p Յ 0.01 for transfected mouse 2B4.11 (2B4.hCD40) and human Jurkat percentage, line 3, or absolute number, line 4). This is true whether (J.hCD40) T cell lines with hCD40 (Fig. 3, A and B; MCF for evaluating the percentage or absolute numbers of CD3ϩCD40ϩ 2B4.hCD40 ϭ 704.66 vs 147.48 for 2B4.11; MCF for J.hCD40 ϭ cells as a part of the whole spleen (Table I, lines 3 and 4) or after 245.72 vs 199.11 for Jurkat) and evaluated the ability of CD40 to enrichment of T lymphocytes (Table I, lines 7 and 8). This phe- activate IL-2 production (Fig. 4, A and B). Interestingly, even with nomenon was not due to an alteration in the splenic T cell popu- PMA/ionomycin stimulation, there was no detectable CD154 ex- lation as a whole, or post-T cell isolation after CII/CFA immuni- pression by either 2B4.hCD40 or J.hCD40 cells (Fig. 3, C and D), zation. There was no significant difference between immunization although we see CD154 expression by Hi5 insect cells infected Downloaded from groups in the percentage (first line) or absolute numbers (second with a baculovirus encoded to express CD154 (data not shown). line) of CD3ϩ T cells, whether evaluating mixed splenocyte, or Because these clones do not express CD154, autocrine CD40 stim- isolated T cell populations (Table I). ulation does not contribute to subsequent findings. We first evaluated the ability of the transfected 2B4.hCD40 and T cell CD40 as a costimulatory molecule J.hCD40 to secrete cytokines in response to CD3 Ϯ CD28 and/or

The findings discussed above raise the possibility that CD40 ex- CD40 stimulation (Fig. 4). To closely mimic the design of our http://www.jimmunol.org/ pressed by T cells may play a role in T cell activation. Experiments mouse ex vivo experiments, in which biologic responses to CD40 presented in Fig. 2 explored whether CD40 engagement could aug- ment CD3 or CD3 plus CD28-mediated cytokine production by splenic T cells isolated from mice immunized with CII/CFA. Be- cause minimal cytokine production was detected from cells stim- ulated with medium alone, isotype control Abs, or anti-CD3 Ab, it is unlikely that the small amounts of residual non-T cells found after T cell enrichment (Fig. 1 and Table I) are APC. As expected, anti-CD3 Ab induced a modest amount of IL-2 (Fig. 2A), and by guest on September 29, 2021 anti-CD28 Ab significantly enhanced CD3-mediated IL-2 produc- tion in all three experimental groups (CD3 vs CD3 plus CD28: naive, p ϭ 0.02; CFA only, p ϭ 0.01; CII/CFA, p Ͻ 0.0001). CD40 stimulation alone did not induce any appreciable IL-2 pro- duction. However, in T cells from mice immunized with CII/CFA, CD40 significantly enhanced the level of CD3 ( p ϭ 0.002)- and CD3 plus CD28 ( p ϭ 0.009)-mediated IL-2 production. This en- hancement did not occur in the largely CD40-negative T cells from naive or CFA-treated mice because there was no significant dif- ference in IL-2 produced between T cells treated with agonists for CD3 vs CD3 plus CD40 or CD3 plus CD28 vs CD3 plus CD28 plus CD40. In addition to IL-2 production, we also evaluated the ability of CD40 to contribute to the production of the proinflammatory cy- tokines IFN-␥ (Fig. 2B) and TNF-␣ (Fig. 2C) by T cells from CIA mice compared with controls. As with IL-2 secretion, CD28 en- hanced CD3-mediated IFN-␥ (Fig. 2B, p Յ 0.001 for all groups) and TNF-␣ (Fig. 2C, p Յ 0.001 for all groups) production in T FIGURE 3. Expression of hCD40 and CD154 on 2B4.11 and Jurkat cell cells from all three mouse groups. Unlike IL-2, anti-CD40 alone lines. A and B, 2B4.11 (A) or Jurkat (B) cells were transfected with DNA induced a significant amount of both IFN-␥ (Fig. 2B, p ϭ 0.001) encoding for hCD40. Cells were stained with anti-human CD40 and ana- and TNF-␣ (Fig. 2C, p Ͻ 0.0001), but only in T cell cultures from lyzed by flow cytometry. Gray profiles represent anti-hCD40 mAb staining mice immunized with CII/CFA and not in cultures from control of untransfected cells; black profiles represent staining of transfected cells. mice. CD40 significantly enhanced CD3 and CD3 plus CD28-me- There was no significant staining of cells with an isotype control Ab. C and diated IFN-␥ (Fig. 2B, p Յ 0.02 for both stimuli) and TNF-␣ ( p Յ D, 2B4.hCD40 (C) or J.hCD40 (D) cells remained untreated or were treated for 6 h with 1 ␮g/ml PMA plus 200 ␮M ionomycin, then stained 0.01 for both stimuli) production, but only in mice immunized with with anti-CD154 and analyzed by flow cytometry. Dashed profiles repre- CII/CFA. These data indicate that CD40 can act as a TCR co- sent staining of untreated cells with isotype control mAb; gray profiles stimulator and that it can cooperate in a nonredundant manner with represent anti-CD154 staining of untreated cells; black profiles represent CD28 to further enhance T cell cytokine production. anti-CD154 staining of PMA-ionomycin-treated cells. There was no sig- The above ex vivo experiments contained a mixed population of nificant staining of PMA-ionomycin-treated cells with an isotype control CD40-expressing and nonexpressing T cells (Fig. 2), with a rela- Ab. Similar results were seen at 24 and 48 h (data not shown). The Journal of Immunology 675

FIGURE 4. CD40 enhances CD3 and CD3 plus CD28-mediated cytokine production in 2B4.hCD40 and J.hCD40 lines. 2B4.11 Ϯ 2B4.hCD40 (A, C, and D) or Jurkat Ϯ J.hCD40 (B) cells were stimulated with plate-bound anti-CD3 Ϯ anti-CD28 and/or anti-CD40 Abs (compared with medium (Med)/isotype control (IC)). Culture supernatants were collected after 24 (TNF-␣), 48 (IL-2), or 72 (IFN-␥) h and assayed for IL-2 (A and B), TNF-␣ (C), or IFN-␥ (D) by ELISA. Data represent the mean Ϯ SEM of triplicate samples from two independent experiments. There was no dif- ference between medium and isotype control samples. Downloaded from

were made by small percentages of CD40ϩ T cells, CD40 trans- The role of T cell CD40 in up-regulation of cell surface fected T cells were cocultured with their untransfected parent cell molecules http://www.jimmunol.org/ lines at various ratios ranging from 6 to 100% transfected cells. CD40 signaling is known to up-regulate a number of cell surface CD40 stimulation alone of cells stably expressing the transfected molecules (27), including CD11␣ (LFA-1), CD54 (ICAM-1), and receptor did not induce IL-2 production (Fig. 4, A and B). While CD95 (Fas), all of which play significant roles in T cell activation CD28 ligation appropriately enhanced CD3-mediated IL-2 produc- (39–41). We also evaluated the ability of CD40 to contribute to tion in all T cell clones, CD40 stimulation enhanced IL-2 produc- CD80 (-1) and CD86 (B7-2) up-regulation, an important com- tion via CD3 or CD3 plus CD28 only if cells stably expressing ponent of activation by other CD40-expressing cells (42), as well CD40 were present in the culture well, similar to responses of as CD25, a marker of T cell activation (43). We compared baseline primary T cell cultures from CIA mice (Fig. 2B). CD40 was able expression of these cell surface molecules on 2B4.11 and ϩ to act in costimulatory fashion with CD3 with only 6% of CD40 2B4.hCD40 cells, as well as receptor-specific up-regulation of by guest on September 29, 2021 cells in the coculture ( p ϭ 0.01 when 6% 2B4.hCD40 present, p Ͻ these molecules postactivation (Fig. 5). Both 2B4.11 and 0.0001 at 100% 2B4.hCD40; p Ͻ 0.001 when 6% J.hCD40 2B4.hCD40 expressed basal CD80 (Fig. 5C), CD86 (Fig. 5D), and present, p Ͻ 0.0001 at 100% J.hCD40), just as was seen in primary CD11␣ (LFA-1; Fig. 5E), with higher expression in 2B4.hCD40 T cells from CIA mice (Fig. 2B). Enhancement of CD3 plus CD28- cells. CD3 plus CD28 stimulation induced up-regulation of all sur- mediated IL-2 production by CD40 increased with greater num- face molecule tested in both 2B4.11 and 2B4.hCD40 cells (right bers of transfected cells in the well ( p ϭ 0.02 at 100% graph in each panel, p Յ 0.01 for all groups). CD40 signals alone, 2B4.hCD40; p ϭ 0.005 when 6% J.hCD40 present, p Ͻ 0.0001 at and in conjunction with CD3 or CD3 plus CD28 stimulation, in- 100% J.hCD40). Of particular interest is that CD40 could ulti- duced up-regulation of all surface molecules tested in 2B4.hCD40 mately enhance CD3-mediated IL-2 production to a degree similar cells, but not the nontransfected 2B4.11 cell line. CD40 signals to that of CD28 in 2B4.hCD40 cells (Fig. 4A) and significantly augmented CD3-mediated up-regulation of cell surface molecules better in J.hCD40 cells (Fig. 4B, p Ͻ 0.0001 at 100% J.hCD40). similar to CD28 stimulation in 2B4.hCD40 cells ( p Ͻ 0.01 CD3 vs Thus, the response of both 2B4.hCD40 and J.hCD40, even when CD3 plus CD28 or CD3 plus CD40 for all groups, no significant ϩ only 6% of the T cells in each culture were CD40 cells, closely difference in response between CD3 plus CD28 and CD3 plus paralleled the responses seen in freshly isolated T cells, validating CD40 stimulation), with a maximal response achieved when triple these cell lines as useful experimental models for the study of CD3 plus CD28 plus CD40 stimulation was given. Interestingly, CD40 function in T cells. CD40 signaling induced a maximal enhancement of the costimu- In addition to IL-2 production, we evaluated the ability of CD40 latory response in 2B4.hCD40 cells in conjunction with CD3 plus to contribute to the production of proinflammatory cytokines CD28 to up-regulate CD25 expression (Fig. 5A). TNF-␣ (Fig. 4C) and IFN-␥ (Fig. 4D). These cytokines were readily detected in culture supernatants of stimulated 2B4.11 or 2B4.hCD40 cells, but not , which have been propagated The role of T cell CD40 in activation of cytokine signaling as an IL-2-producing human T cell line (24) and which pathways require overexpression of other proteins to induce TNF-␣ (37) and Experiments presented above demonstrate that CD40 can act as a IFN-␥ (38) secretion. Unlike IL-2, CD40 stimulation alone was costimulatory molecule to enhance CD3 and CD3 plus CD28-me- able to stimulate both production of TNF-␣ ( p ϭ 0.02 when 6% diated T cell activation. NFAT, AP-1, and NF-␬B are known to be 2B4.hCD40 cells were present, p ϭ 0.001 at 100%) and IFN-␥ involved in the activation of several T cell proinflammatory cyto- ( p ϭ 0.02 when 6% 2B4.hCD40 cells were present, p Ͻ 0.0001 at kine genes, including those encoding IL-2 (44–46), IFN-␥ (47, 100%) in those cells stably expressing CD40. Similar to IL-2, 48), and TNF-␣ (45, 49). We and others have demonstrated that CD40 stimulation of T cells expressing CD40 was able to augment CD40-mediated activation of B lymphocytes involves AP-1 (50, both CD3 and CD3 plus CD28-mediated cytokine production. 51) and NF-␬B (31, 52) signaling pathways, and there is evidence 676 SIGNALING BY CD40 IN T CELLS

FIGURE 5. CD40 enhances CD3 and CD3 plus CD28-mediated cell surface molecule up-regulation. 2B4.11 or 2B4.hCD40 cells were stimulated with plate-bound anti-

CD3 Ϯ anti-CD28 and/or anti-CD40 Downloaded from Abs (compared with isotype control) for 48–72 h, then analyzed by flow cytometry to determine expression levels of surface proteins. Histograms (gray ϭ isotype control; black ϭ sur- face molecule) represent baseline ex- pression (stimulated with isotype con- http://www.jimmunol.org/ trol Ab), with corresponding median channel fluorescence (⌬MCF ϭ MCF of surface molecule Ϫ MCF of iso- type control staining) represented in the left graph of each panel. Changes in MCF for each surface molecule tested (poststimulation minus base- line) is represented in the right graph of each panel. A–F, Data represent the by guest on September 29, 2021 mean Ϯ SEM of MCF from two in- dependent experiments.

of CD40-mediated NFAT activation (51, 53). We asked if this is pared with CD3 plus CD40 in 2B4.hCD40 cells, p ϭ 0.017 compared also the case for T cell CD40. with CD3 plus CD28, p ϭ 0.003 compared with CD3 plus CD40 in CD3 or CD40 signals alone induced minimal NFAT reporter J.hCD40 cells). Although CD3 signals alone were able to trigger in- gene activation compared with control stimuli in 2B4.hCD40 cells creased NFAT activity in J.hCD40 cells ( p ϭ 0.001; Fig. 6B), CD28 (Fig. 6A), while CD28 in conjunction with CD3 signals induced an and CD40 signals augmented this response in a manner similar to increased response ( p ϭ 0.001) compared with medium/isotype their effects in 2B4.hCD40 cells. controls or CD3 single stimulation in 2B4.hCD40 cells (Fig. 6A; We next investigated the ability of CD40 to signal via AP-1 in p Ͻ 0.0001 compared with medium/isotype controls or CD3 in T cells (Fig. 6). Unlike NFAT (Fig. 6), CD40 signals alone acti- J.hCD40 cells, Fig. 6B). While CD40 signaling was able to en- vated AP-1 ϳ2-fold over control stimuli in 2B4.hCD40 cells (Fig. hance CD3-mediated NFAT activation ( p Ͻ 0.001 in both 7A; p ϭ 0.005). While CD40 signals enhanced CD3-mediated 2B4.hCD40 and J.hCD40 cells), it was not to the same degree as AP-1 activation in both cell lines ( p Ͻ 0.001 in 2B4.hCD40 cells; CD28-mediated enhancement. The greatest NFAT response was p ϭ 0.009 in J.hCD40 cells), CD40 was a less efficient costimu- achieved via engagement of all three receptors: CD3 plus CD28 plus lator in this response than CD28 (CD3 vs CD3 plus CD28: p ϭ CD40 ( p ϭ 0.007 compared with CD3 plus CD28, p Ͻ 0.001 com- 0.007 in 2B4.hCD40 cells; p ϭ 0.005 in J.hCD40), and the The Journal of Immunology 677

FIGURE 6. CD40 enhances CD3 and CD3 plus CD28-mediated NFAT activation. 2B4.hCD40 (A)or J.hCD40 (B) cells were transiently transfected with 4ϫ NFAT-luciferase and Renilla-luciferase reporter plas- mids, rested on ice for 30 min, then stimulated for 24 h. Cells were stimulated with anti-hCD40 or isotype con- trol (IC) Abs or Dynal beads armed with anti-CD3 or anti-CD3 plus anti-CD28 Abs. Relative luciferase ac- tivity (NFAT:Renilla) of stimulus vs control was cal- culated as the mean Ϯ SEM of duplicate samples from three independent experiments. There was no difference between medium and isotype control samples. Downloaded from http://www.jimmunol.org/ maximal response was again achieved with simultaneous engage- CD40 was engaged was strong phosphorylation of JNK observed, ment of CD3 plus CD28 plus CD40 ( p ϭ 0.02 compared with CD3 with no phosphorylation seen via CD3 signaling and minimal plus CD28, p ϭ 0.03 compared with CD3 plus CD40 in phosphorylation via CD3 plus CD28 signaling. 2B4.hCD40 cells, p ϭ 0.01 compared with CD3 plus CD28, p ϭ It is well established that CD40 signals lead to NF-␬B activation in 0.002 compared with CD3 plus CD40 in J.hCD40 cells). CD40 B cells (52, 56–58) and we asked if this was also true for T cell CD40. signals in B cells strongly activate phosphorylation of the AP-1 CD40 alone initiated a strong NF-␬B response in 2B4.hCD40 (Fig. family member c-jun, via activation of JNK (54–56). We evalu- 9A, p Ͻ 0.001) and J.hCD40 (Fig. 9B, p Ͻ 0.001) cells, while CD3 ated the ability of T cell CD40 to contribute to JNK activation evoked a minimal response compared with controls. CD3 plus CD40 alone or in combination with CD3 or CD3 plus CD28. Strikingly, activated NF-␬B more than CD40 alone (CD40 vs CD3 plus CD40: by guest on September 29, 2021 in both 2B4.hCD40 and J.hCD40 cells (Fig. 8, A and B), only when p ϭ 0.03 in 2B4.hCD40 cells, p ϭ 0.02 in J.hCD40 cells) and 2.5- to 5-fold greater than CD3 plus CD28 (CD3 plus CD28 vs CD3 plus CD40: p ϭ 0.005 in 2B4.hCD40 cells; p ϭ 0.004 in J.hCD40 cells), while the combination of CD3 plus CD28 plus CD40 signals provided an additional 20% increase in NF-␬B activation ( p Ͻ 0.001 compared with CD3 plus CD28, p ϭ 0.03 compared with CD3 plus CD40 in 2B4.hCD40 cells; p ϭ 0.01 compared with CD3 plus CD28, p ϭ 0.04 compared with CD3 plus CD40 in J.hCD40 cells). Early events in NF-␬B activation have been shown to include phosphorylation and degradation of I␬B␣ (59), and recent studies suggest that an alternate pathway for activating NF-␬B (NF-␬B2), in which p100 is processed to p52 and shuttled to the nucleus by RelB, is also used by some TNFR family members, including CD40 (56, 58). To test which NF-␬B activation pathways are used by CD40 in T cells, we assayed for I␬B␣ phosphorylation and degradation (NF-␬B1; Fig. 10) or processing of p100 to p52 with nuclear translocation of p52 and RelB (NF-␬B2; Fig. 11). Using densitometry to normalize I␬B␣ values to the loading control actin (Fig. 10B), CD40 engagement alone stimulated phosphorylation and degradation of I␬B␣ with up to 100% greater efficiency than CD3 or CD3 plus CD28, with an even greater increase when CD40 FIGURE 7. CD40 activates and enhances CD3 and CD3 plus CD28- signals were combined with CD3 or CD3 plus CD28. Similar re- mediated AP-1 activation. 2B4.hCD40 (A) or J.hCD40 (B) cells were tran- sults were seen in J.hCD40 cells (data not shown). With respect to ϫ siently transfected with 7 AP-1-luciferase and Renilla-luciferase reporter the NF-␬B2 pathway, in both 2B4.hCD40 (Fig. 11) and J.hCD40 plasmids, rested on ice for 30 min, then stimulated for 24 h. Cells were (data not shown), CD40 stimulation alone, but not CD3, resulted in stimulated with anti-hCD40 or isotype control (IC) Abs or Dynal beads armed with anti-CD3 or anti-CD3/anti-CD28 Abs. Relative luciferase ac- processing of p100 to p52 (Fig. 11, A and B) and translocation of tivity (AP-1:Renilla) of stimulus vs control cell groups was calculated as p52 and RelB to the nucleus (Fig. 11, C and D). Using densitom- the mean Ϯ SEM of duplicate samples from two independent experiments. etry to normalize p100, p52, and RelB values to the loading control There was no difference between medium and isotype control samples. actin (Fig. 11B, cytoplasmic fraction) or yy1 (Fig. 11D, nuclear 678 SIGNALING BY CD40 IN T CELLS

FIGURE 8. CD40 activates and enhances CD3 and CD3 plus CD28- mediated JNK phosphorylation. 2B4.hCD40 (A) or J.hCD40 (B) cells were rested at 37°C for 30 min, then stimulated for 5–60 min with anti-hCD40 or isotype control (IC) Abs or Dynal beads armed with anti- CD3 or anti-CD3 plus anti-CD28 Abs. Cells were pelleted, lysed, and lysates analyzed by SDS-PAGE and Western blot for phosphorylated (P.) and total (T.) JNK. Similar results were obtained in two independent experiments.

fraction), we observed that CD40 and CD28 augmented CD3-me- TRAF2 and TRAF3 moved efficiently into the lipid raft fraction (left diated activation of NF-␬B2 to a similar degree, whereas the com- panel) and associated with CD40 (right panel) in both 2B4.hCD40

bination of CD3 plus CD28 plus CD40 gave maximal stimulation. (Fig. 12A) and J.hCD40 (Fig. 12B) cells, similar to their recruitment Downloaded from in B cells (61). This was true if hCD40 was engaged by agonistic Abs TRAF association of T cell CD40 or by CD154, although more efficient movement (left panel) and re- CD40, like other members of the TNFR superfamily, relies on the ceptor association (right panel) of TRAF1 and TRAF6 occurred when association of adaptor molecules, TRAFs, for downstream signal- hCD154 was the stimulus, as previously reported (63, 64). ing events, including activation of and transcription fac-

tors, production of cytokines, up-regulation of surface molecules, Discussion http://www.jimmunol.org/ and various aspects of the humoral response (20). However, the All B cells express CD40, and its functions on B cells have been characteristics of TRAF association with CD40 have been shown the subject of much study. Much less well appreciated is that ac- to differ between B cells, macrophages, dendritic cells, and epi- tivated T cells can also express CD40, and the roles and mecha- thelial cells (19). It was thus important to determine CD40-TRAF nisms of action of T cell-expressed CD40 still represent a signif- associations in T cells. In mouse B cells, we have previously dem- icant knowledge gap. Various TNFR family members have been onstrated that TRAFs 1, 2, 3, and 6 associate with either endoge- shown to act as coregulatory molecules on T cells, some possibly nous mouse CD40 or transfected hCD40, following receptor liga- contributing to inflammatory disease (65). This role has been pre- tion (19, 60–62). We therefore compared the ability of TRAFs to viously established for CD40 on B cells (66), and we demonstrate move into membrane lipid rafts (Fig. 12, A and B, left panel) and here that this is also true for T cell CD40. We see a reproducible by guest on September 29, 2021 associate with hCD40 (Fig. 12, A and B, right panel) in Brij sol- increase in CD40 expression on T cells from mice with the chronic uble (cytoplasmic) and insoluble (lipid raft) fractions in T cells. inflammatory disorder CIA (Fig. 1 and Table I). Although the in- crease in CD40ϩ cells in CIA vs control mice is modest compared with genetically predisposed autoimmune-prone strains of mice (14, 17), it is consistent with the FACS profile of T cells from normal mice (9, 17, 67). Strains of mice that are prone to sponta- neous have robust expression of CD40 by T cells, whereas nondisease prone strains do not (14), so it is not surprising that C57BL/6 mice do not have as robust proportions of CD40ϩ T cells as strains that develop autoimmune diseases rela- tively early in life. A normal mouse strain can induce functional CD40 on T cells when encountering an Ag/danger signal (67), but this level still does not reach that of mouse strains prone to spon- taneous autoimmune disease, which have likely been receiving chronic stimulation since early in life. However, findings presented here show that T cell CD40 can contribute to enhanced T cell activation even in individuals who do not have strong genetic pre- disposition to autoimmunity. Importantly, this CD40 is capable of T cell stimulation, indicat- ing that it may have significant functional consequences (9, 14, 15, 67). Initially, this was a surprising finding. However, the Ag-spe- cific precursor frequency for naive T cells is ϳ1/1 ϫ 105 (100 FIGURE 9. CD40 activates and enhances CD3 and CD3 plus CD28- cells/spleen) (68, 69), a small population capable of significantly ␬ mediated NF- B activation. 2B4.hCD40 (A) or J.hCD40 (B) cells were expanding and producing a sufficient protective adaptive immune transiently transfected with 4ϫ NF-␬B-luciferase and Renilla-luciferase response upon antigenic stimulation. This response ultimately reporter plasmids, rested for 15 min on ice, then stimulated for 6 h. Cells ϳ ϳ ϫ 5 were stimulated with anti-hCD40 or isotype control (IC) Abs or Dynal leads to the survival of 5% of the activated T cells ( 1 10 beads armed with anti-CD3 or anti-CD3 plus anti-CD28 Abs. Relative cells) to serve as a memory pool after the contraction phase (69). luciferase activity (NF-␬B:Renilla) of stimulus vs control was calculated as Interestingly, in the CIA model, at 70 days postimmunization, we 5 ϩ the mean Ϯ SEM of duplicate samples from two independent experiments. can isolate a population of 3–4 ϫ 10 CD40 T cells per spleen There was no difference between medium and isotype control samples. (Table I) that are capable of CD40-mediated activation of cytokine The Journal of Immunology 679

FIGURE 10. CD40 activates and enhances I␬B␣ phosphorylation/degradation. A, 2B4.hCD40 cells were stimulated with anti-hCD40 or isotype control (IC) Abs or Dynal beads armed with anti-CD3 or anti-CD3 plus anti-CD28 Abs. Cells were pelleted, lysed, and lysates analyzed by SDS-PAGE and Western blot for phos- phorylated and total I␬B␣. B, Density of bands as a proportion of cells treated with medium (med) alone and normalized to the value of the actin bands shown in B. Similar results were obtained in two independent experiments.

ϳ ϩ production (Fig. 2), despite only representing 7% of the isolated percentage of CD40 cells as that observed in CIA mice. Impor- Downloaded from T cell population (Table I). This finding was recapitulated in ex- tantly, anti-CD40 stimulation does not yield a positive cytokine periments whereby CD40ϩ T cells were mixed with CD40Ϫ T response in T cells lacking CD40, although their response to CD3 cells in a controlled fashion and a significant CD40-mediated re- plus CD28 stimulation is similar to that of CD40ϩ T cells (Figs. 2 sponse was seen with as few as 6% CD40ϩ T cells in culture and 4). Activation of the transcription factor NFAT is critical to (Fig. 4). IL-2 production (44). Also consistent with a role as a costimulator, Like other costimulatory TNFR family members expressed on T CD40 itself cannot induce NFAT activation (Fig. 6), but can aug- http://www.jimmunol.org/ cells (70), while CD40 itself cannot induce IL-2 production, it ment these responses to CD3 and CD3 plus CD28 ligation. It is augments the CD3 response and gives maximal stimulation to- likely that the TCR complex provides calcium-mediated signaling gether with CD3 plus CD28 signals (Figs. 2 and 4). This is true in that is necessary, but not sufficient, for T cell activation and IL-2 both T cells from CIA mice (Fig. 2) and in T cell lines expressing production (71), while CD40 provides costimulatory signaling via CD40 (Fig. 4, A and B), even when transfected CD40ϩ cells are NF-␬B and AP-1, necessary to activate NFAT and subsequent IL-2 mixed with CD40Ϫ (untransfected) T cells to give a similar small production (44, 72, 73). by guest on September 29, 2021

FIGURE 11. CD40 activates and enhances CD3 and CD3 plus CD28-mediated nuclear translocation of p52 and RelB in the NF-␬B2 pathway. 2B4.hCD40 (A–D) cells were stimulated with anti-hCD40 or isotype con- trol (IC) Abs or Dynal beads armed with anti-CD3 or anti-CD3/anti-CD28 Abs. Cells were pelleted, cyto- plasmic and nuclear fractions isolated, and lysates an- alyzed by SDS-PAGE and Western blot for p100/p52 and RelB. Blots were stripped and reblotted for actin (cytoplasmic fractions; A and B) or yy1 (nuclear frac- tions; C and D) as loading controls. B, Density of bands as a proportion of cells treated with medium (med) alone and normalized to the value of the actin bands shown in A. D, Density of bands as a proportion of cells treated with medium (med) alone, normalized to the value of the yy1 bands shown in C. Similar results were obtained in two independent experiments. 680 SIGNALING BY CD40 IN T CELLS

B cells in its pathogenesis (84), and CD40 maybe a potent co- stimulator in this process. CD40 expressed by B and T cells may use similar molecular mechanisms, described above, to contribute to the pathogenesis of inflammation in autoimmune disease. CD40 on both B cells (18, 85–89) and T cells (this study) synergizes with AgRs to enhance activation, cytokine production (Figs. 2 and 4), and up-regulation of cell surface molecules (Fig. 5). CD40 signaling leads to isotype switching and autoantibody production in B cells (90). In T cells, it has been demonstrated that CD40 engagement leads to TCR revision within germinal centers (67), skewing the T cell population further toward autoimmunity (14–16). A proinflammatory cytokine environment is critical for reaching the threshold of autoimmune disease development (65, 91). CD40 engagement in either T or B cells leads to TNF-␣ secretion, as shown in Figs. 2C and 4C and (56, 92). We have previously re- ported that CD40 signaling to B cells is partially mediated by FIGURE 12. CD40-mediated TRAF recruitment. 2B4.hCD40 (A)or TNF-␣ binding to TNFR2 (56, 92) and hypothesize this also to be

J.hCD40 (B) cells were stimulated for 15 min with anti-hCD40 Ab (vs true of T cell CD40, as it has been demonstrated that TNF-␣ acts Downloaded from isotype control Ab; IC) or Hi5 insect cells expressing hCD154 (vs WT via TNFR2 to lower the threshold of T cell activation and IL-2 baculovirus; WT). Detergent soluble (S) and insoluble (I) lysates were production (93–95). Taken together, the presence of increased prepared as described in Materials and Methods. Seventy percent of the numbers of CD40ϩ T cells in autoimmune mice and the demon- lysates were immunoprecipitated with protein G-Sepharose armed with stration that CD40 can act as an effective costimulatory receptor on anti-hCD40. Lysates and immunoprecipitates were separated by SDS- T cells suggest that blockade of CD40-CD154 interactions can PAGE and analyzed by Western blot. Similar results were obtained in two

abrogate the pathogenesis of autoimmune disease on several lev- http://www.jimmunol.org/ identical experiments. els. This has been demonstrated in the experimental autoimmune encephalomyelitis model of multiple sclerosis, whereby disease Both CD40 (74) and CD28 (75) contribute to cell activation via development is dependent on CD40 signaling, particularly in the association with lipid rafts. CD40 signals use TRAF adaptor mol- absence of CD28 (96). In the HgCl2-induced autoimmunity model, ecules in B cells (61, 64), and we demonstrate in Fig. 12 that the CD28 is unable to overcome the lack of CD40 signaling to induce same is true for T cell CD40. TRAFs 1, 2, 3, and 6 bind CD40 disease (97). within the Brij insoluble (raft) fraction upon engagement with ag- CD40 may not just be a TCR costimulatory molecule on T cells onistic anti-CD40 Ab or membrane-bound CD154. This suggests but may make these cells effective APC by up-regulating cell sur- that TRAFs are a critical component of T cell CD40 signaling, face molecules (Fig. 5). Increased levels of not only CD40, but by guest on September 29, 2021 providing a key difference from CD28-mediated signaling. As in B other costimulatory molecules such as CD80, CD86, ICAM-1, and cells (50), T cell CD40 efficiently activates up-regulation of cell LFA-1, may also be pivotal in autoimmune disease development surface molecules (Fig. 5), both canonical and noncanonical (97). Although beyond the scope of this study, it would be inter- NF-␬B pathways (Figs. 10 and 11), AP-1 (Fig. 7), and the AP-1 esting to investigate the costimulatory capacity of activated CD40- activator JNK (Fig. 8). CD40 as a costimulatory molecule is as expressing T cells for CD40-negative T cells, as well as for other effective as CD28 at signaling via AP-1 and severalfold more ef- cell types, including APC. It is quite possible that activation of ficient at signaling via NF-␬B, with maximal increase in both re- CD40 on T cells lowers the threshold of disease development but sponses when stimulating T cells with agonists for CD3 plus CD28 is not sufficient for it to occur, requiring CD40 activation on APC plus CD40. This suggests that CD40 and CD28 may have different for cytokine production and on B cells for autoantibody secretion. molecular mechanisms leading to activation of AP-1 and NF-␬B, In this way, CD40 would prove central to autoimmune disease as has been proposed when comparing CD28 and other TNFR development and pathogenesis, influencing not only the cognitive family members on T cells (76). Importantly, it shows that CD40 activation of APC but also T cells. As seen in Fig. 2, results in can provide more powerful enhancement of NF-␬B, a transcription enhanced effector proinflammatory cytokine production that along factor that induces cytokine genes with particular potency, than with CD40 activation on B cells would lead to autoantibody pro- other T cell costimulators. This is also seen in the activation of duction and, ultimately, chronic inflammation. JNK, which in the two T cell lines tested, was seen only when a CD40 costimulus was included. Disclosures The above findings suggest that CD40 can increase the potency The authors have no financial conflict of interest. and number of signaling pathways available to T cells that express it. This is important when considering threshold requirements for References 1. Grewal, I. S., and R. A. Flavell. 1996. The role of CD40 ligand in costimulation developing autoimmune disease and associated chronic inflamma- and T cell activation. Immunol. Rev. 153: 85–106. tion. The presence of CD40-expressing T cells or autoantibodies is 2. Rodriguez-Pinto, D., and J. Moreno. 2005. B cells can prime naive CD4ϩ T cells not enough to develop autoimmunity. NOR mice, like their NOD in vivo in the absence of other professional -presenting cells in a CD154- CD40-dependent manner. Eur. J. Immunol. 35: 1097–1105. counterparts, have CD40-expressing T cells, yet do not develop 3. Bishop, G. A., and B. S. Hostager. 2003. The CD40-CD154 interaction in B disease (17). Similarly, the presence of autoantibodies does not cell-T cell liaisons. Cytokine Growth Factor Rev. 14: 297–309. necessarily indicate pathogenesis (77–79). Cooperation between 4. Balasa, B., T. Krahl, G. Patstone, J. Lee, R. Tisch, H. O. McDevitt, and N. Sarvetnick. 1997. CD40 ligand-CD40 interactions are necessary for the initi- cell types and signals from the environment to the immune re- ation of and diabetes in nonobese diabetic mice. J. Immunol. 159: sponse determine development of disease (80, 81). 4620–4627. 5. Durie, F. H., R. A. Fava, T. M. Foy, A. Aruffo, J. A. Ledbetter, and R. J. Noelle. Like many autoimmune diseases, including diabetes (82), ar- 1993. Prevention of collagen-induced arthritis with an to gp39, the thritis (83), and lupus (7), CIA requires cooperation between T and ligand for CD40. Science 261: 1328–1330. The Journal of Immunology 681

6. Toubi, E., and Y. Shoenfeld. 2004. The role of CD40-CD154 interactions in 37. Ballester, A., A. Velasco, R. Tobena, and S. Alemany. 1998. Cot kinase activates autoimmunity and the benefit of disrupting this pathway. Autoimmunity 37: tumor necrosis factor ␣ in a cyclosporin A-resistant manner. 457–464. J. Biol. Chem. 273: 14099–14106. 7. Datta, S. K. 1998. Production of pathogenic : cognate interactions be- 38. Kashiwakura, J., N. Suzuki, H. Nagafuchi, M. Takeno, Y. Takeba, tween autoimmune T and B cells. Lupus 7: 591–596. Y. Shimoyama, and T. Sakane. 1999. Txk, a nonreceptor kinase of the 8. Kyburz, D., M. Corr, D. C. Brinson, A. Von Damm, H. Tighe, and D. A. Carson. Tec family, is expressed in T helper type 1 cells and regulates interferon ␥ pro- 1999. Human rheumatoid factor production is dependent on CD40 signaling and duction in human T lymphocytes. J. Exp. Med. 190: 1147–1154. autoantigen. J. Immunol. 163: 3116–3122. 39. Chung, D. H., K. C. Jung, W. S. Park, I. S. Lee, W. J. Choi, C. J. Kim, S. H. Park, 9. Fanslow, W. C., K. N. Clifford, M. Seaman, M. R. Alderson, M. K. Spriggs, and Y. Bae. 2000. Costimulatory effect of Fas in mouse T lymphocytes. Mol. Cell R. J. Armitage, and F. Ramsdell. 1994. Recombinant CD40 ligand exerts potent 10: 642–646. biologic effects on T cells. J. Immunol. 152: 4262–4269. 40. Tibbetts, S. A., C. Chirathaworn, M. Nakashima, D. S. Jois, T. J. Siahaan, 10. Bourgeois, C., B. Rocha, and C. Tanchot. 2002. A role for CD40 expression on M. A. Chan, and S. H. Benedict. 1999. Peptides derived from ICAM-1 and CD8ϩ T cells in the generation of CD8ϩ T cell memory. Science 297: LFA-1 modulate T cell adhesion and immune function in a mixed lymphocyte 2060–2063. culture. Transplantation 68: 685–692. 11. Koschella, M., D. Voehringer, and H. Pircher. 2004. CD40 ligation in vivo in- 41. Moulian, N., J. Bidault, C. Planche, and S. Berrih-Aknin. 1998. Two signaling duces bystander proliferation of memory phenotype CD8 T cells. J. Immunol. pathways can increase fas expression in human . Blood 92: 172: 4804–4811. 1297–1307. 12. Lee, B. O., L. Hartson, and T. D. Randall. 2003. CD40-deficient, influenza- 42. Bhatia, S., M. Edidin, S. C. Almo, and S. G. Nathenson. 2006. B7-1 and B7-2: specific CD8 memory T cells develop and function normally in a CD40-sufficient similar costimulatory ligands with different biochemical, oligomeric and signal- environment. J. Exp. Med. 198: 1759–1764. ing properties. Immunol. Lett. 104: 70–75. 13. Sun, J. C., and M. J. Bevan. 2004. Cutting edge: long-lived CD8 memory and 43. Adler, S. H., E. Chiffoleau, L. Xu, N. M. Dalton, J. M. Burg, A. D. Wells, protective immunity in the absence of CD40 expression on CD8 T cells. J. Im- M. S. Wolfe, L. A. Turka, and W. S. Pear. 2003. Notch signaling augments T cell munol. 172: 3385–3389. responsiveness by enhancing CD25 expression. J. Immunol. 171: 2896–2903. 14. Wagner, D. H., Jr., E. Newell, R. J. Sanderson, J. H. Freed, and M. K. Newell. 44. Cianferoni, A., M. Massaad, S. Feske, M. A. de la Fuente, L. Gallego,

1999. Increased expression of CD40 on thymocytes and peripheral T cells in N. Ramesh, and R. S. Geha. 2005. Defective nuclear translocation of nuclear Downloaded from autoimmunity: a mechanism for acquiring changes in the peripheral T cell re- factor of activated T cells and extracellular signal-regulated kinase underlies de- ceptor repertoire. Int. J. Mol. Med. 4: 231–242. ficient IL-2 gene expression in Wiskott-Aldrich syndrome. J. Allergy Clin. Im- 15. Wagner, D. H., Jr., G. Vaitaitis, R. Sanderson, M. Poulin, C. Dobbs, and munol. 116: 1364–1371. K. Haskins. 2002. Expression of CD40 identifies a unique pathogenic T cell 45. Palanki, M. S. 2002. Inhibitors of AP-1 and NF-␬B mediated transcriptional population in . Proc. Natl. Acad. Sci. USA 99: 3782–3787. activation: therapeutic potential in autoimmune diseases and structural diversity. 16. Vaitaitis, G. M., M. Poulin, R. J. Sanderson, K. Haskins, and D. H. Wagner Jr. Curr. Med. Chem. 9: 219–227. 2003. Cutting edge: CD40-induced expression of recombination activating gene 46. Jain, J., C. Loh, and A. Rao. 1995. Transcriptional regulation of the IL-2 gene. (RAG) 1 and RAG2: a mechanism for the generation of autoaggressive T cells in Curr. Opin. Immunol. 7: 333–342. the periphery. J. Immunol. 170: 3455–3459. 47. Kaminuma, O., C. Elly, Y. Tanaka, A. Mori, Y. C. Liu, A. Altman, and http://www.jimmunol.org/ 17. Waid, D. M., G. M. Vaitaitis, and D. H. Wagner, Jr. 2004. Peripheral S. Miyatake. 2002. Vav-induced activation of the human IFN-␥ gene promoter is CD4loCD40ϩ auto-aggressive T cell expansion during insulin-dependent diabetes mediated by upregulation of AP-1 activity. FEBS Lett. 514: 153–158. mellitus. Eur. J. Immunol. 34: 1488–1497. 48. Sica, A., L. Dorman, V. Viggiano, M. Cippitelli, P. Ghosh, N. Rice, and 18. Haxhinasto, S. A., B. S. Hostager, and G. A. Bishop. 2002. Cutting edge: mo- H. A. Young. 1997. Interaction of NF-␬B and NFAT with the interferon ␥ pro- lecular mechanisms of synergy between CD40 and the B cell antigen receptor: moter. J. Biol. Chem. 272: 30412–30420. role for TNF receptor-associated factor 2 in receptor interaction. J. Immunol. 169: 49. Li-Weber, M., M. K. Treiber, M. Giaisi, K. Palfi, N. Stephan, S. Parg, and 1145–1149. P. H. Krammer. 2005. Ultraviolet irradiation suppresses T cell activation via 19. Bishop, G. A., and B. S. Hostager. 2001. Molecular mechanisms of CD40 sig- blocking TCR-mediated ERK and NF-␬B signaling pathways. J. Immunol. 175: naling. Arch. Immunol. Ther. Exp. 49: 129–137. 2132–2143. 20. Bishop, G. A., and B. S. Hostager. 2001. Signaling by CD40 and its mimics in 50. Baccam, M., S. Y. Woo, C. Vinson, and G. A. Bishop. 2003. CD40-mediated

B cell activation. Immunol. Res. 24: 97–109. transcriptional regulation of the IL-6 gene in B lymphocytes: involvement of by guest on September 29, 2021 21. Croft, M. 2003. Co-stimulatory members of the TNFR family: keys to effective NF-␬B, AP-1, and C/EBP. J. Immunol. 170: 3099–3108. T cell immunity? Nat. Rev. Immunol. 3: 609–620. 51. Francis, D. A., J. G. Karras, X. Y. Ke, R. Sen, and T. L. Rothstein. 1995. In- 22. Watts, T. H. 2005. TNF/TNFR family members in costimulation of T cell re- duction of the transcription factors NF-␬B, AP-1 and NF-AT during B cell stim- sponses. Ann. Rev. Immunol. 23: 23–68. ulation through the CD40 receptor. Int. Immunol. 7: 151–161. 23. Ashwell, J. D., R. E. Cunningham, P. D. Noguchi, and D. Hernandez. 1987. Cell 52. Hsing, Y., and G. A. Bishop. 1999. Requirement for nuclear factor ␬B activation growth cycle block of T cell hybridomas upon activation with antigen. J. Exp. by a distinct subset of CD40-mediated effector functions in B lymphocytes. J. Im- Med. 165: 173–194. munol. 162: 2804–2811. 24. Gillis, S., and J. Watson. 1980. Biochemical and biological characterization of 53. Antony, P., J. B. Petro, G. Carlesso, N. P. Shinners, J. Lowe, and W. N. Khan. lymphocyte regulatory molecules. V. Identification of an 2-producing 2003. B cell receptor directs the activation of NFAT and NF-␬B via distinct human leukemia T cell line. J. Exp. Med. 152: 1709–1719. molecular mechanisms. Exp. Cell Res. 291: 11–24. 25. Warren, W. D., and M. T. Berton. 1995. Induction of germ-line ␥1 and ␧ Ig gene 54. Brown, K. D., B. S. Hostager, and G. A. Bishop. 2001. Differential signaling and expression in murine B cells—IL-4 and the CD40 ligand-CD40 interaction pro- tumor necrosis factor receptor-associated factor (TRAF) degradation mediated by vide distinct but synergistic signals. J. Immunol. 155: 5637–5646. CD40 and the Epstein-Barr virus oncoprotein latent membrane protein 1 (LMP1). 26. Bishop, G. A., W. D. Warren, and M. T. Berton. 1995. Signaling via major J. Exp. Med. 193: 943–954. histocompatibility complex class II molecules and antigen receptors enhances the 55. Li, Y. Y., M. Baccam, S. B. Waters, J. E. Pessin, G. A. Bishop, and B cell response to gp39/CD40 ligand. Eur. J. Immunol. 25: 1230–1238. G. A. Koretzky. 1996. CD40 ligation results in protein kinase C-independent 27. Hostager, B. S., Y. Hsing, D. E. Harms, and G. A. Bishop. 1996. Different activation of ERK and JNK in resting murine splenic B cells. J. Immunol. 157: CD40-mediated signaling events require distinct CD40 structural features. J. Im- 1440–1447. munol. 157: 1047–1053. 56. Munroe, M. E., and G. A. Bishop. 2004. Role of tumor necrosis factor (TNF) 28. Bishop, G. A., and J. A. Frelinger. 1989. Haplotype-specific differences in sig- receptor-associated factor 2 (TRAF2) in distinct and overlapping CD40 and TNF naling by transfected class II molecules to a Ly-1ϩ B cell clone. Proc. Natl. Acad. receptor 2/CD120b-mediated B lymphocyte activation. J. Biol. Chem. 279: Sci. USA 86: 5933–5937. 53222–53231. 29. Campbell, I. K., J. A. Hamilton, and I. P. Wicks. 2000. Collagen-induced arthritis 57. Hsing, Y., B. S. Hostager, and G. A. Bishop. 1997. Characterization of CD40 in C57BL/6 (H-2b) mice: new insights into an important disease model of rheu- signaling determinants regulating nuclear factor ␬B activation in B lymphocytes. matoid arthritis. Eur. J. Immunol. 30: 1568–1575. J. Immunol. 159: 4898–4906. 30. Rosloniec, E. F., A. H. Kang, L. K. Myers, and M. A. Cremer. 1997. Collagen- 58. Coope, H. J., P. G. Atkinson, B. Huhse, M. Belich, J. Janzen, M. J. Holman, Induced Arthritis. Wiley & Sons, New York. G. G. Klaus, L. H. Johnston, and S. C. Ley. 2002. CD40 regulates the processing 31. Berberich, I., G. L. Shu, and E. A. Clark. 1994. Cross-linking CD40 on B cells of NF-␬B2 p100 to p52. EMBO J. 21: 5375–5385. rapidly activates nuclear factor ␬B. J. Immunol. 153: 4357–4366. 59. Karin, M., and Y. Ben-Neriah. 2000. Phosphorylation meets ubiquitination: the 32. Schreiber, E., P. Matthias, M. M. Muller, and W. Schaffner. 1989. Rapid detec- control of NF-␬B activity. Ann. Rev. Immunol. 18: 621–663. tion of octamer binding proteins with “mini-extracts,” prepared from a small 60. Bishop, G. A., B. S. Hostager, and K. D. Brown. 2002. Mechanisms of TNF number of cells. Nucleic Acids Res. 17: 6419. receptor-associated factor (TRAF) regulation in B lymphocytes. J. Leukocyte 33. Moore, C. R., and G. A. Bishop. 2005. Differential regulation of CD40-mediated Biol. 72: 19–23. TNF receptor-associated factor degradation in B lymphocytes. J. Immunol. 175: 61. Hostager, B. S., I. M. Catlett, and G. A. Bishop. 2000. Recruitment of CD40 and 3780–3789. tumor necrosis factor receptor-associated factors 2 and 3 to membrane microdo- 34. Xie, P., B. S. Hostager, and G. A. Bishop. 2004. Requirement for TRAF3 in mains during CD40 signaling. J. Biol. Chem. 275: 15392–15398. signaling by LMP1 but not CD40 in B lymphocytes. J. Exp. Med. 199: 661–671. 62. Jalukar, S. V., B. S. Hostager, and G. A. Bishop. 2000. Characterization of the 35. Myers, L. K., E. F. Rosloniec, M. A. Cremer, and A. H. Kang. 1997. Collagen- roles of TNF receptor-associated factor 6 in CD40-mediated B lymphocyte ef- induced arthritis, an animal model of autoimmunity. Life Sci. 61: 1861–1878. fector functions. J. Immunol. 164: 623–630. 36. Schulze-Koops, H., and J. R. Kalden. 2001. The balance of Th1/Th2 cytokines in 63. Hostager, B. S., S. A. Haxhinasto, S. L. Rowland, and G. A. Bishop. 2003. Tumor . Best Pract. Res. Clin. Rheumatol. 15: 677–691. necrosis factor receptor-associated factor 2 (TRAF2)-deficient B lymphocytes 682 SIGNALING BY CD40 IN T CELLS

reveal novel roles for TRAF2 in CD40 signaling. J. Biol. Chem. 278: 81. Shepshelovich, D., and Y. Shoenfeld. 2006. Prediction and prevention of auto- 45382–453890. immune diseases: additional aspects of the mosaic of autoimmunity. Lupus 15: 64. Xie, P., B. S. Hostager, M. E. Munroe, C. R. Moore, and G. A. Bishop. 2006. 183–190. Cooperation between TNF receptor-associated factors 1 and 2 in CD40 signaling. 82. Noorchashm, H., S. A. Greeley, and A. Naji. 2003. The role of t/b lymphocyte J. Immunol. 176: 5388–5400. collaboration in the regulation of autoimmune and alloimmune responses. Immu- 65. Kwon, B., B. S. Kim, H. R. Cho, J. E. Park, and B. S. Kwon. 2003. Involvement nol. Res. 27: 443–450. of tumor necrosis factor receptor superfamily (TNFRSF) members in the patho- 83. Korganow, A. S., H. Ji, S. Mangialaio, V. Duchatelle, R. Pelanda, T. Martin, genesis of inflammatory diseases. Exp. Mol. Med. 35: 8–16. C. Degott, H. Kikutani, K. Rajewsky, J. L. Pasquali, et al. 1999. From systemic 66. Lorenz, H. M., M. Herrmann, and J. R. Kalden. 2001. The pathogenesis of au- T cell self-reactivity to organ-specific autoimmune disease via immunoglobulins. toimmune diseases. Scand. J. Clin. Lab. Invest. 235: 16–26. Immunity 10: 451–461. 67. Cooper, C. J., G. L. Turk, M. Sun, A. G. Farr, and P. J. Fink. 2004. Cutting edge: 84. Brand, D. D., A. H. Kang, and E. F. Rosloniec. 2003. Immunopathogenesis of TCR revision occurs in germinal centers. J. Immunol. 173: 6532–6536. collagen arthritis. Springer Semin. Immunopathol. 25: 3–18. 68. Whitmire, J. K., N. Benning, and J. L. Whitton. 2006. Precursor frequency, non- ϩ 85. Haxhinasto, S. A., and G. A. Bishop. 2003. Synergistic B cell activation by CD40 linear proliferation, and functional maturation of virus-specific CD4 T cells. and the B cell antigen receptor: role of BCR-mediated kinase activation and J. Immunol. 176: 3028–3036. TRAF regulation. J. Biol. Chem. 279: 2575–2582. 69. Blattman, J. N., R. Antia, D. J. Sourdive, X. Wang, S. M. Kaech, K. Murali-Krishna, J. D. Altman, and R. Ahmed. 2002. Estimating the precursor 86. Haxhinasto, S. A., and G. A. Bishop. 2003. A novel interaction between protein frequency of naive antigen-specific CD8 T cells. J. Exp. Med. 195: 657–664. kinase D and TNF receptor-associated factor molecules regulates B cell receptor- 70. Saoulli, K., S. Y. Lee, J. L. Cannons, W. C. Yeh, A. Santana, M. D. Goldstein, CD40 synergy. J. Immunol. 171: 4655–4662. N. Bangia, M. A. DeBenedette, T. W. Mak, Y. Choi, and T. H. Watts. 1998. 87. Mizuno, T., and T. L. Rothstein. 2003. Cutting edge: CD40 engagement elimi- CD28-independent, TRAF2-dependent costimulation of resting T cells by 4-1BB nates the need for Bruton’s tyrosine kinase in B cell receptor signaling for NF- ␬ ligand. J. Exp. Med. 187: 1849–1862. B. J. Immunol. 170: 2806–2810. 71. Gupta, S. 1989. Mechanisms of transmembrane signalling in human T cell acti- 88. Mizuno, T., and T. L. Rothstein. 2005. B cell receptor (BCR) cross-talk: CD40 vation. Mol. Cell. Biochem. 91: 45–50. engagement creates an alternate pathway for BCR signaling that activates I␬B 72. Flescher, E., J. A. Ledbetter, N. Ogawa, N. Vela-Roch, D. Fossum, H. Dang, and kinase/I␬B␣/NF-␬B without the need for PI3K and phospholipase C␥. J. Immu- N. Talal. 1995. Induction of transcription factors in human T lymphocytes by nol. 174: 6062–6070. Downloaded from aspirin-like drugs. Cell. Immunol. 160: 232–239. 89. Mizuno, T., and T. L. Rothstein. 2005. B cell receptor (BCR) cross-talk: CD40 73. Isakov, N., and A. Altman. 2002. Protein kinase C␪ in T cell activation. Ann. Rev. engagement enhances BCR-induced ERK activation. J. Immunol. 174: Immunol. 20: 761–794. 3369–3376. 74. Kaykas, A., K. Worringer, and B. Sugden. 2001. CD40 and LMP-1 both signal 90. Ollila, J., and M. Vihinen. 2005. B cells. Int. J. Biochem. Cell Biol. 37: 518–523. from lipid rafts but LMP-1 assembles a distinct, more efficient signaling complex. 91. Kollias, G., and D. Kontoyiannis. 2002. Role of TNF/TNFR in autoimmunity: EMBO J. 20: 2641–2654. specific TNF receptor blockade may be advantageous to anti-TNF treatments. 75. Kovacs, B., R. V. Parry, Z. Ma, E. Fan, D. K. Shivers, B. A. Freiberg, Cytokine Growth Factor Rev. 13: 315–321.

A. K. Thomas, R. Rutherford, C. A. Rumbley, J. L. Riley, and T. H. Finkel. 2005. 92. Hostager, B. S., and G. A. Bishop. 2002. Role of TNF receptor-associated factor http://www.jimmunol.org/ Ligation of CD28 by its natural ligand CD86 in the absence of TCR stimulation 2 in the activation of IgM secretion by CD40 and CD120b. J. Immunol. 168: induces lipid raft polarization in human CD4 T cells. J. Immunol. 175: 3318–3322. 7848–7854. 93. Kim, E. Y., and H. S. Teh. 2004. Critical role of TNF receptor type-2 (p75) as a 76. Watts, T. H., and M. A. DeBenedette. 1999. T cell co-stimulatory molecules other costimulator for IL-2 induction and T cell survival: a functional link to CD28. than CD28. Curr. Opinion. Immunol. 11: 286–293. J. Immunol. 173: 4500–4509. 77. Nowak, U. M., and M. M. Newkirk. 2005. Rheumatoid factors: good or bad for you? Int. Arch. Allergy Immunol. 138: 180–188. 94. Kim, E. Y., and H. S. Teh. 2001. TNF type 2 receptor (p75) lowers the threshold 78. Schlosser, M., K. Koczwara, H. Kenk, M. Strebelow, I. Rjasanowski, of T cell activation. J. Immunol. 167: 6812–6820. R. Wassmuth, P. Achenbach, A. G. Ziegler, and E. Bonifacio. 2005. In insulin- 95. Aspalter, R. M., M. M. Eibl, and H. M. Wolf. 2003. Regulation of TCR-mediated autoantibody-positive children from the general population, antibody affinity T cell activation by TNF-RII. J. Leukocyte Biol. 74: 572–582. identifies those at high and low risk. Diabetologia 48: 1830–1832. 96. Girvin, A. M., M. C. Dal Canto, and S. D. Miller. 2002. CD40/CD40L interaction

79. Quintana, F. J., and I. R. Cohen. 2004. The natural autoantibody repertoire and is essential for the induction of EAE in the absence of CD28-mediated co-stim- by guest on September 29, 2021 autoimmune disease. Biomed. Pharmacother. 58: 276–281. ulation. J. Autoimmun. 18: 83–94. 80. Matsui, K., S. Jodo, S. Xiao, and S. T. Ju. 1999. Hypothesis: a recurrent, mod- 97. Pollard, K. M., M. Arnush, P. Hultman, and D. H. Kono. 2004. Costimulation erate activation fosters systemic autoimmunity—the apoptotic roles of TCR, IL-2 requirements of induced murine systemic autoimmune disease. J. Immunol. 173: and . J. Biomed. Sci. 6: 306–313. 5880–5887.