Subsets Immunological Synapse in CD4 T Cell CTLA-4 Differentially

CTLA-4 Differentially Regulates the Immunological Synapse in CD4 T Cell Subsets

This information is current as Rachael P. Jackman, Fran Balamuth and Kim Bottomly of September 24, 2021. J Immunol 2007; 178:5543-5551; ; doi: 10.4049/jimmunol.178.9.5543 http://www.jimmunol.org/content/178/9/5543 Downloaded from

References This article cites 52 articles, 29 of which you can access for free at: http://www.jimmunol.org/content/178/9/5543.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 24, 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

CTLA-4 Differentially Regulates the Immunological Synapse in CD4 T Cell Subsets1

Rachael P. Jackman,2 Fran Balamuth, and Kim Bottomly

Primary murine Th1 and Th2 cells differ in the organization of the immunological synapse, with Th1 cells, but not Th2 cells, clustering signaling molecules at the T cell/B cell synapse site. We sought to determine whether differential costimulatory signals could account for the differences observed. We found that Th2 cells express higher levels of CTLA-4 than Th1 cells, and demonstrated that Th2 cells lacking CTLA-4 are now able to cluster the TCR with the same frequency as Th1 cells. Furthermore, reconstitution of CTLA-4 into CTLA-4-deﬁcient Th2 cells, or into Th1 cells, inhibits the clustering of the TCR. We have also shown that Th2 cells, but not Th1 cells, show variations in the organization of the immunological synapse depending on levels of expression of CD80/CD86 on the APC. These studies demonstrate a unique role for CTLA-4 as a critical regulator of Th2 cells

and the immunological synapse. The Journal of Immunology, 2007, 178: 5543–5551. Downloaded from

ctivation of T cells requires recognition of a peptide/ the absence of PKC␪ clustering and that CTLs can kill without MHC by a speciﬁc, unique TCR. In addition to this re- substantial clustering of the TCR (11, 12). Studies in this lab- A quirement, a number of other costimulatory molecules oratory have shown using primary cells that Th1 cells cluster help to regulate activation. Varying expression and organization of PKC␪, lipid rafts, and the TCR when stimulated with resting signaling molecules at the cell surface is a critical way of regulat-

splenic B cells and specific peptide, but Th2 cells do not (13). http://www.jimmunol.org/ ing what is needed to activate a T cell. By generating, degrading, These studies suggest that the formation and the organization of sequestering, or colocalizing different signaling molecules, T cells the immunological synapse are not simply an on/off switch for are able to fine-tune their activation thresholds. activation, but that they may regulate signaling in more subtle One way in which this may be accomplished is through the ways. To date, the function of the immunological synapse reorganization of the immunological synapse. The immunological mains unclear, but understanding how synapse organization is synapse forms at the site of contact between a T cell and its cog- regulated will help elucidate the role of these highly ordered nate APC. The classical immunological synapse is described as a structures. central clustering of small signaling molecules, such as the TCR, 3 Mechanisms controlling the clustering of signaling molecules CD28, and protein kinase C␪ (PCK␪), surrounded by a ring of by guest on September 24, 2021 and lipid rafts and the organization of the synapse site have been larger adhesion molecules (1). This structure forms following early examined, and many factors have been shown to play a role. Our TCR signaling, and has been proposed to be important for sus- laboratory has shown that CD4 plays a critical role in regulating taining activation signals by segregating activating and inhibitory the clustering of the TCR, PKC␪, and lipid rafts in the synapse site signaling molecules, for promoting close contact between cells by through a mechanism dependent on palmitoylation of CD4 and displacing large surface molecules at the site of interaction, as well as for attenuating signaling by down-regulation of the TCR (2–8). interactions with lck, but that this mechanism is independent of Differences in immunological synapse organization have been ob- those controlling microtubule-organizing center reorientation and served, suggesting that not all T cell/APC conjugates organize talin clustering (13, 14). CD28 and CTLA-4 are important regu- their synapses in the same way. Immature thymocytes have been lators of T cell activation, and have also been shown to participate shown in bilayer experiments to form multiple, small, transient in synapse composition. CD28 is expressed constitutively on the clusters of the TCR, but have no central clustering of the TCR or surface of most CD4 T cells (15), and has been shown to cluster at lipid rafts (9, 10). It has also been observed that naive CD8 T cells the site of contact with the APC, where it helps to recruit lipid rafts can be activated to proliferate, differentiate, and produce IL-2 in (16–18). In contrast, CTLA-4 is up-regulated following activation, and has been shown to shut down strong signals by relocalizing to the site of contact and entering lipid rafts, where it appears to Yale University School of Medicine, Department of Immunobiology, New Haven, ␨ CT 06520 reduce levels of phosphorylated TCR -chain in lipid rafts (19, 20). Both of these molecules bind CD80/CD86 molecules on the sur- Received for publication May 22, 2006. Accepted for publication February 14, 2007. face of the interacting APC, with CD28 acting as a critical co- The costs of publication of this article were defrayed in part by the payment of page stimulator required for naive T cell activation (15), and CTLA-4 charges. This article must therefore be hereby marked advertisement in accordance acting as an inhibitor of T cell activation. Levels and kinetics of with 18 U.S.C. Section 1734 solely to indicate this fact. expression of CD28, CTLA-4, CD80, and CD86, their localization 1 This work was supported by National Institutes of Health Grants CA38350, A126791, and T32A107019. within the cell, and the differing affinities and avidities of CTLA-4 and CD28 for their ligands all impact which signaling response 2 Address correspondence and reprint requests to Dr. Rachael Jackman, Yale School of Medicine, Department of Immunobiology, TAC 540, P.O. Box 208011, will dominate (21–23). Thus, both CD28 and CLTA-4 appear to New Haven, CT 06520. E-mail address: [email protected] play some role in determining the organization of the immunolog- 3 Abbreviations used in this paper: PKC␪, protein kinase C␪; BMDC, bone marrow- ical synapse, and their shared ligand and opposing signaling out- derived dendritic cell; DC, dendritic cell; MCC, moth cytochrome c. comes make them potentially important candidates for regulating Copyright © 2007 by The American Association of Immunologists, Inc. 0022-1767/07/$2.00 the type of synapse structure that forms. www.jimmunol.org 5544 CTLA-4 AND THE CD4 T CELL IMMUNOLOGICAL SYNAPSE

In these studies, we sought to understand the mechanism con- B cell activation trolling the differential organization of the immunological synapse T cell-depleted splenic cells were treated with 30 ␮g/ml LPS for 72 h and in Th1 and Th2 cells. Specifically, we asked whether differential used to restimulate effector T cells, as described above. costimulatory signals, particularly CD28 and CTLA-4, could be responsible for the observed differences in synapse organization in CD80/CD86 blocking these two effector subsets. We have shown in this study that Th2 T cell-depleted APC were incubated with a mixture of anti-CD80 and anti- cells express higher levels of CTLA-4 than Th1 cells and that this CD86 Abs at a concentration of 5 ␮g/ml (unless specified) for 15 min at room increased CTLA-4 expression is responsible for the observed dif- temperature, following peptide pulse and before use in stimulation of T cells. ferences in synapse organization. In the absence of CTLA-4, Th2 Immunocytochemistry and microscopy cells that fail to cluster the TCR upon activation in the presence T cells were mixed with peptide-pulsed APC, as described above, for 10 of CTLA-4 now cluster the TCR with the same frequency as min at 37°C. T cell/APC conjugates were then placed on Alcian blue- Th1 cells. Furthermore, reconstitution of CTLA-4 into CTLA- coated coverslips in serum-free medium and incubated for 30 min at 37°C 4-deficient Th2 cells, or into Th1 cells, inhibited the clustering to permit adherence. Cells were fixed in ethanol, permeabilized with sa- of the TCR. In addition, our data indicate that Th2 cells, unlike ponin, and stained, as previously described (28). Analysis was done using a conventional Zeiss microscope and the Openlab software (Improvision). Th1 cells, vary the organization of the immunological synapse depending on the levels of CD80/CD86 expression on the APC, Flow cytometry analysis suggesting that the type of APC, or the degree to which that For intracellular cytokine staining, cells were restimulated for 4 h with 0.1 APC is activated, might determine the type of synapse that is ␮g/ml PMA and 0.75 ␮g/ml ionomycin in the presence of GolgiStop and

formed. permeabilized using the BD Cytoﬁx/Cytoperm kit following their standard Downloaded from protocol (BD Biosciences). Ethidium monoazide was used to distinguish Materials and Methods and exclude dead cells (Invitrogen Life Technologies). Antibodies Retroviral transduction Anti-CD4 (GK1.5), anti-CD8 (TIB 210), anti-Thy-1 (Y19), anti-class II CTLA-4 cDNA was purchased from American Type Culture Collection MHC (212.A1, TIB 92, or 14.4.4), anti-CD32/16 (24G2), anti-IFN-␥ (accession BC042741), and modiﬁed by PCR to introduce BglII (5Ј) and

(XMG1.2), and anti-B220 (TIB 164) were purified from culture superna- XhoI(3Ј) restriction sites to subclone into pMSCV retroviral expression http://www.jimmunol.org/ tants on protein G columns and dialyzed against PBS before use. Biotin and vector (provided by K. Murphy, Washington University, St. Louis, MO), FITC anti-V␤3 (KJ25), purified anti-CTLA-4 (UC10-4F10-11), PE anti- which contains a cassette with GFP and an internal ribosome entry site IFN-␥ (XMG1.2), allophycocyanin anti-IL-4 (11B11), FITC anti-V␣11 permitting the translation of two open reading frames from one mRNA. (RR8-1), biotin and FITC anti-B220 (RA3-6B2), allophycocyanin anti- The plasmids containing CTLA-4-internal ribosome entry site-GFP or GFP CD11c (HL3), PE anti-I-Ek (14-4-4S), PE anti-ICAM-1 (3E2), FITC alone were transfected into Phoenix Ecotropic packaging cells (provided anti-CD80 (16-10A1), purified anti-CD80 (1G10/B7), purified and by G. Nolan, Stanford University, Stanford, CA), and 24 h later, were FITC anti-CD86 (GL1), PE anti-CD8b.2 (53-5.8), PerCP anti-CD4 transferred to 30°C for 48 h to allow viral production. The retroviral su- (RM4-5), and biotin anti-hamster IgG (G70-204 and G94-56) were pur- pernatants were used to spin infect Th1 or Th2 effector cells (2000 rpm chased from BD Pharmingen; purified anti-talin (C20) was purchased for 90 min) 48 h following restimulation with peptide and APCs. Cells from Santa Cruz Biotechnology. Alexa-488 anti-FITC, streptavidin were then incubated at 37°C for 96 h, rested for 48 h, and sorted for GFP-positive cells using the FACSVantage SE cell sorter (BD Bio- by guest on September 24, 2021 Alexa-488, Alexa-594 anti-goat, Alexa-647 anti-mouse, and streptavi- ϩ din Alexa-647 were purchased from Molecular Probes. sciences). Approximately 10% of cells were GFP . Cells were analyzed for CTLA-4 expression by flow cytometry. Sorted GFPϩ cells were then Mice restimulated with MCC-pulsed syngeneic APC and analyzed by fluorescence microscopy, as described above. B10.BR mice were obtained from The Jackson Laboratory. The AND TCR transgenic mice have been previously described (24). C57BL/6 CTLA- Generation of Th1 and Th2 cells in vivo 4Ϫ/Ϫ, CD80Ϫ/Ϫ, and CD86Ϫ/Ϫ mice were received from A. Sharpe (Har- vard Medical School, Boston, MA). These mice were crossed onto B6 Sensitization and challenge were performed, as previously described (29– AND TCR transgenic and B6 Rag2Ϫ/Ϫ Ϫ/Ϫ 31). For intranasal sensitization, mice were lightly anesthetized with isoflu- mice to generate B6 Rag AND ␮ TCR Tg CTLA-4Ϫ/Ϫ Ϫ/Ϫ ϩ/Ϫ rane and subjected to intranasal delivery of 100 g of OVA protein (grade and B6 Rag AND TCR Tg CTLA-4 mice. All ␮ mice used to generate CD4 T cells in these studies were 6–10 wk old. All V; Sigma-Aldrich) in 50 l of PBS on days 0, 1, and 2, as previously ␮ animal protocols were approved by Yale’s Institutional Animal Care and described. For low-dose LPS treatment, 0.05 gofEscherichia coli LPS Use Committee 2004-10393. O55:B5 (Sigma-Aldrich) was used on each day of sensitization. For high- dose LPS treatment, 20 ␮g of LPS was used. All mice were lightly anes- Peptides thetized with isoflurane and challenged intranasally with 25 ␮g of OVA in 50 ␮l of PBS on days 14, 15, 18, and 19, and sacrificed on day 21. Me- The peptide used in these studies is derived from moth cytochrome c diastinal lymph node cells were isolated and allowed to rest in culture for (MCC; peptide 81–103), pMCC ϭ VFAGLKKANERADLIAYLKQATK. 5 days. These cells were then restimulated with PMA and ionomycin in the Peptide was synthesized and purified by HPLC by the W. M. Keck Foun- presence of GolgiStop for 4 h, then fixed, permeabilized, and stained for dation Biotechnology Resource Laboratory. intracellular IL-4, IFN-␥, and CTLA-4, as described above. Preparation of APCs, CD4ϩ T cells, and bone marrow-derived Statistical analysis dendritic cells (BMDCs) To demonstrate significance of TCR clustering quantitation, 95% confi- T cell-depleted APC were prepared by Ab-mediated complement lysis of dence intervals were calculated and are shown as error bars. Nonoverlap- ϩ Ϫ ␣ ϭ B10.BR splenocytes, as previously described (25). CD4 CD8 T cells ping error bars indicate significance at 0.05 level. Values of p were from lymph nodes and spleens of transgenic mice were isolated using im- calculated using binomial distribution analysis. For conjugate quantitation, munomagnetic negative selection, as previously described (25). BMDCs a minimum of 10 fields and 50 conjugates was counted for each group. were generated from B10.BR bone marrow cultured with GM-CSF, as previously described (26). These cells were cultured without further stimuli Results and used on day 7, by which time the cells have begun to mature. Th1 and Th2 cells differ in immunological synapse organization Generation and activation of effector T cells As shown in Fig. 1, and as we have previously published, unlike Th1 cells, Th2 cells do not cluster the TCR, CD4, and PKC␪ at the Th1 and Th2 cells were generated by in vitro cytokine skewing, as previously described (27). Effector T cells were restimulated by coculture with site of T cell/APC contact when stimulated with primary B cells T-depleted splenocytes from B10.BR mice pulsed with 50 ␮M indicated and peptide (13). In these studies, Th1 and Th2 cells were prepared peptide, as previously described (13). from AND TCR transgenic mice and tested for their ability to The Journal of Immunology 5545

the synapse site (Fig. 1, B and C). In our splenic APC preparations, these events are rare (only ϳ2% of the cells are B220Ϫ and class II MHCϩ), making it difﬁcult to gather quantitative information on synapses with these cells. Therefore, we asked whether Th2 cells could cluster the TCR at the T cell/APC interface upon activation with other APCs, such as dendritic cells (DCs). To address the ability of Th2/DC conjugates to aggregate the TCR at the conjugation site, we stimulated Th1 and Th2 cells with peptide-pulsed BMDCs and examined conjugates between CD4 T cells and class II MHC-positive BMDCs. Similar to what we observed with T cell/B cell conjugates, both Th1 and Th2 cells clustered talin at the T/DC contact site (Fig. 2A). In contrast to the results observed with resting B cells, the TCR (V␤3) localized to the T/DC contact site with both Th1 and Th2 cells (Fig. 2, A and B). As has been seen previously, it was common to see multiple T cells actively conju- gated with a single DC in which all or many of the T cells had clustered the TCR at the site of contact (36). Because other studies have shown clustering of signaling mol-

ecules at the CD4 T cell/APC site between Th2 clones and B cell Downloaded from lymphomas (1), we asked whether our primary Th2 cells would cluster the TCR when stimulated with activated B cells. B cells were activated with LPS for 72 h, pulsed with peptide, and then used to stimulate Th1 and Th2 cells. Under these conditions, unlike what was seen with resting B cells, both Th1 and Th2 cells

clustered the TCR at the CD4 T cell/APC contact site with the http://www.jimmunol.org/ same frequency (Fig. 2, C and D). Therefore, although Th2 cells do not cluster the TCR at the conjugation site with resting B cells, they are able to do so when stimulated with either mature BMDCs or activated B cells. CD80/CD86 regulate synapse organization We observed aggregation of the TCR at the synapse site in Th2 FIGURE 1. Th1 and Th2 cells differ in their organization of the immu- cells only when stimulated with peptide presented by activated B nological synapse. Th1 and Th2 cells were generated by cytokine skewing cells and BMDCs, but not resting B cells. Because CD80/CD86 by guest on September 24, 2021 of CD4 T cells from AND TCR transgenic mice, as previously described interactions with CD28 and CTLA-4 have been shown to be im- (27), and were stimulated with peptide-pulsed APCs. T cell/APC conju- portant in T cell activation and cell membrane organization, we gates were stained with anti-V␤3, anti-talin, and anti-B220. TCR and talin asked whether differential CD80/CD86 expression might be re- ␤ ϩ ϩ localization in T cell (V 3 )/B cell (B220 ) conjugates (A) and T cell/ sponsible for the observed difference in TCR clustering in Th1 and B220Ϫ cell conjugates (B). C, Conjugates that clustered talin at the T cell/B Ϫ Th2 cells. In keeping with previous accounts, we observed that cell or T cell/B220 cell synapse site were examined for TCR clustering. Shown here is the percentage of these conjugates in which the TCR clusters both resting and activated B cells, as well as activated BMDCs, express high levels of class II MHC and ICAM-1, but differ in at the T cell/APC synapse site. Difference in TCR-clustering frequency Ϫ between Th1 and Th2 cells is significant with p Ͻ 0.0001. Error bars mark CD80/CD86 expression, with activated B cells, B220 spleno- ϩ 95% confidence intervals. For all experiments, a minimum of 10 fields and cytes, and class II MHC BMDCs expressing higher levels of 50 active T cell/APC conjugates was counted. This experiment was re- both CD80 and CD86 than resting B cells (data not shown). Be- peated more than five times. cause costimulatory molecules have been shown previously to be required for TCR clustering at the site of conjugation in naive T cluster the TCR, PKC␪, and talin at the T cell/APC interface. Talin cells (17, 18, 37), we looked to see whether CD80/CD86 was is a molecule that activates LFA and links it to the cytoskeleton, required for TCR clustering in Th1 or Th2 cells. T cell-depleted and is used in this study as a functional indicator of active conju- splenocytes were cultured with LPS for 72 h, pulsed with peptide, gation between the T cell and its APC (32–35). Of the conjugates then preincubated with either a mixture of anti-CD80 and anti- ϩ ϩ between V␤3 and B220 cells defined by clustered talin, only CD86 Abs or isotype control Abs before being used to stimulate T 17% of Th2 cells had clustered TCR as compared with 55% of Th1 cells. Active T cell/B cell conjugates were examined by micros- cells (Fig. 1, A and C). There was no difference in the number of copy for clustering of the TCR. As seen in Fig. 3A, both Th1 and conjugates observed between Th1 and Th2 cells, with both subsets Th2 cells clustered the TCR at the site of contact, as expected, clustering talin at the T cell/B cell synapse site with equal fre- when the activated B cells were preincubated with an irrelevant quency (data not shown). Similarly, PKC␪ only clustered in Th1/ isotype control Ab (54% of Th1 cells and 53% of Th2 cells). When APC conjugates, not Th2/APC conjugates (data not shown). anti-CD80 and anti-CD86 Abs were used to block the interaction Therefore, whereas both Th1 and Th2 cells are able to form active of CD80 and CD86 with their ligands, the frequency of cells in conjugates with resting B cells, only Th1 cells are able to cluster which clustering of the TCR at the site of conjugation was ob- the TCR and PKC␪ under these conditions. served was significantly reduced for both Th1 and Th2 cells (Fig. 3B). This shows that costimulatory molecules (CD80/CD86) are Th2 cell synapse organization varies with APC type required for efficient aggregation of the TCR at the conjugation site We have observed that B220Ϫ APCs from the spleen appear to be in both Th1 and Th2 effector cells. Furthermore, these data suggest capable of signaling both Th1 and Th2 cells to cluster the TCR at that whereas Th1 cells are able to cluster the TCR at the site of 5546 CTLA-4 AND THE CD4 T CELL IMMUNOLOGICAL SYNAPSE

FIGURE 2. Both Th1 and Th2 cells cluster the TCR at the synapse site with DCs and LPS-activated B cells. Th1 and Th2 cells were generated as above and were stimulated with Downloaded from peptide-pulsed BMDCs (A). T cell/BMDC conjugates were stained with anti-V␤3, anti- talin, and anti-I-Ek. B, Quantitation of TCR clustering at the T cell/BMDC site (as described in Fig. 1C) representing two experiments. C, Untreated or LPS-activated B cells were used to stimulate Th1 and Th2 cells. D, http://www.jimmunol.org/ Quantitation of TCR clustering at the T cell/B cell site using LPS-activated vs resting B cells (as described in Fig. 1C). This experiment was repeated more than ﬁve times. by guest on September 24, 2021

conjugation when stimulated with APCs that express low levels of TCR clustering, Th2 cells require a higher expression level of CD80/CD86, Th2 cells require higher levels of costimulation from these costimulatory molecules than Th1 cells. their cognate APCs. To confirm that Th2 cells require higher expression of CD80/ Th2 cells express higher levels of CTLA-4 CD86 on their cognate APC than Th1 cells to cluster the TCR at Because Th1 cells have a lower threshold than Th2 cells for the T cell/B cell synapse site, CD80/CD86 availability was varied CD80/CD86 signals required for TCR clustering, we investi- by titrating the blocking CD80/CD86 Abs in vitro. LPS-stimulated gated how CD80/CD86 might be playing a differential role in B cells were preincubated with a mixture of anti-CD80 and anti- immunological synapse organization. Because CD80 and CD86 CD86 Abs at either 5, 0.01, or 0 ␮g/ml. Active T cell/B cell con- interact with both CD28 and CTLA-4, we examined expression jugates were examined by microscopy for clustering of the TCR, of CD28 and CTLA-4 in Th1 and Th2 cells. CD28 levels were as described above. As was observed in Fig. 3B, in the absence of comparable in Th1 and Th2 cells (data not shown). CTLA-4 has blocking Abs, both Th1 and Th2 cells clustered the TCR at the T been shown to rapidly relocalize to the cell surface (20), so we cell/B cell synapse site, and in the presence of the highest concen- examined CTLA-4 levels by intracellular staining as surface tration of blocking Abs, neither subset clustered the TCR effi- levels alone might under-represent available CTLA-4 protein. ciently (Fig. 3C). At the intermediate concentration of blocking As seen in Fig. 4A, CTLA-4 expression was significantly higher Ab, Th1 cells maintained clustering at the synapse site, but Th2 in resting Th2 cells (right panel) compared with resting Th1 cells did not (Fig. 3C). This confirms that whereas both Th1 and cells (left panel). The same differences were observed following Th2 cells require CD80/CD86 on their cognate APC for optimal multiple rounds of stimulation (data not shown). The Journal of Immunology 5547 Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 3. CD80/CD86 molecules are critical in regulating the type of synapse formed. A, CD80/CD86 interactions were blocked on LPS-activated B cells using anti-CD80 and anti-CD86 Abs or isotype control Abs. These cells were used to activate Th1 and Th2 cells. B, Quantitation of TCR clustering at the T cell/B cell site with disruption of CD80/CD86 binding (as described in Fig. 1C). Difference in TCR-clustering frequency FIGURE 4. Th2 cells express higher levels of CTLA-4. A, Intracellular with isotype control vs blocking Abs is significant for both Th1 and Th2 stain of resting Th1 and Th2 cells for CTLA-4 (gating on live T cells); Ͻ cells with p 0.0001. This experiment was repeated more than five times. isotype control in gray. B, Surface stain of resting Th1 and Th2 cells for C, A titration of CD80/CD86-blocking Abs was used to examine partial CTLA-4; isotype control in gray. C, Relative mean fluorescent intensity of block of CD80/CD86. B cells were incubated with anti-CD80 and anti- surface (black) and total (gray) CTLA-4 normalized to matched isotype ␮ CD86 Abs at 0, 0.01, or 5 g/ml concentrations, and then used to stimulate control. D, Intracellular stain for IL-4 in PMA/ionomycin-restimulated Th1 Th1 and Th2 cells. Quantitation of TCR clustering at the T cell/B cell site and Th2 cells. Gates for IL-4ϩ and IL-4Ϫ populations used for analysis in with partial block of CD80/CD86 (as described in Fig. 1C). Difference in C. E, Intracellular CTLA-4 levels in IL-4ϩ and IL-4Ϫ T cells were com- ␮ TCR-clustering frequency between Th1 and Th2 cells with 0.01 g/ml Ab pared (0.4% of Th1 cells and 27.3% of Th2 cells). block is significant with p Ͻ 0.0001.

whereas no CTLA-4 was observed on the cell surface in Th1 cells The expression of CTLA-4 on the cell surface has been shown (left panel). Relative mean ﬂuorescent intensities of both surface to be tightly regulated, with CTLA-4 localized primarily in intra- and total CTLA-4 were calculated, expressed as a fold increase cellular compartments in resting T cells (38–42). To determine over the values for the matched isotype control (Fig. 4C). These whether any of the CTLA-4 present in Th1 or Th2 cells was on the values suggest that ϳ25% of the total CTLA-4 present in Th2 cells cell surface, nonpermeabilized Th1 or Th2 cells were also stained is expressed on the cell surface. Furthermore, the small amount of for CTLA-4. As seen in Fig. 4B, resting Th2 cells constitutively CTLA-4 expressed in Th1 cells appears to be exclusively expressed low levels of CTLA-4 on the cell surface (right panel), intracellular. 5548 CTLA-4 AND THE CD4 T CELL IMMUNOLOGICAL SYNAPSE Downloaded from

FIGURE 5. Th2 cells generated in vivo express higher levels of http://www.jimmunol.org/ CTLA-4. Mice were intranasally sensitized with OVA protein and either high- or low-dose LPS on days 0, 1, and 2; challenged intranasally with OVA on days 14, 15, 18, and 19; and sacriﬁced on day 21. Mediastinal lymph node cells were isolated and allowed to rest in vitro for 5 days. These cells were pooled, restimulated with PMA and ionomycin, and stained intracellularly for IL-4, IFN-␥, and CTLA-4. A, Expression of IFN-␥ and IL-4 in CD4ϩ cells. Isotype controls for these stains are shown on the left. Gates show cells used to compare CTLA-4 expression in B. B, CTLA-4 (right panel) and isotype control (left panel) stains of the IFN-␥ϩ by guest on September 24, 2021 and IL-4ϩ cells.

To confirm that we were looking at Th2 cells, and not contaminating Th1 or undifferentiated CD4 T cells, we examined levels of CTLA-4 expression in Th2 cells as defined by IL-4 expression. To accomplish this, we restimulated the Th1 cells (Fig. 4D, left panel) and Th2 cells (Fig. 4D, right panel) for 4 h and examined levels of FIGURE 6. CTLA-4-deficient Th1 and Th2 cells cluster the TCR at the CTLA-4 and IL-4 intracellularly. Gating on IL-4-producing Th2 synapse site with resting B cells. A, Th1 and Th2 cells from CTLA-4ϩ/Ϫ cells (Fig. 4D, right panel), we analyzed expression of CTLA-4 on (upper two panels) and CTLA-4Ϫ/Ϫ (lower two panels) mice were stimu- those cells (Fig. 4E, right panel). We observed enrichment for lated with peptide-pulsed APCs. B, Quantitation of TCR clustering at the CTLA-4 expression on the IL-4ϩ cells (Fig. 4E, right panel). This T cell/B cell site in the presence and absence of CTLA-4 (as described in Ϫ Ϫ is in comparison with gating on the non-IL-4-producing cells (Fig. Fig. 1C). Difference in TCR-clustering frequency between CTLA-4 / ϩ/Ϫ 4E, right panel, IL-4Ϫ), which expressed little CTLA-4. Therefore, Th2 cells vs CTLA-4 Th2 cells is significant with p Ͻ 0.0001. This in considering Th2 cells defined by IL-4 production, Th2 cells experiment was repeated more than five times. express higher levels of CTLA-4 than non-IL-4-producing cells. Th1 cells express very little IL-4 or CTLA-4. However, gating on the contaminating IL-4-producing cells from the Th1-skewed cells PMA/ionomycin before examination for IFN-␥, IL-4, and CTLA-4 (only 0.4% of the cells (Fig. 4D, left panel)) demonstrated that expression by flow cytometry. IL-4-producing cells were com- these cells also express high levels of CTLA-4 (Fig. 4E, left panel). pared with IFN-␥-producing cells for CTLA-4 expression. As can Taken together, these data clearly demonstrate higher expression be seen in Fig. 5, the IL-4-producing cells expressed higher levels of CTLA-4 in Th2 cells than in Th1 cells. of CTLA-4 than the IFN-␥-producing cells. These results clearly To confirm our finding of higher expression of CTLA-4 in Th2 demonstrate that Th2 effector cells generated in vivo express cells in vivo, we used a murine model of asthma using high- and higher levels of CTLA-4 than Th1 effector cells. low-dose LPS during sensitization to skew the immune response toward either a Th1 or a Th2 response, as previously described CTLA-4 regulates Th2 cell synapse (29–31). Briefly, BALB/c mice were intranasally sensitized with To investigate whether the increased expression of CTLA-4 in Th2 OVA and either a high or low dose of LPS, and then challenged 2 cells interfered with the clustering of the TCR when stimulated wk later with intranasal OVA. Cells were harvested from the me- with resting B cells, we generated Th1 and Th2 cells from CTLA- diastinal lymph nodes and allowed to rest in vitro for 5 days; then 4-deficient TCR transgenic mice. CTLA-4-deficient mice develop the high- and low-dose groups were pooled and restimulated with lymphoproliferative disease if left untreated (43, 44), but CTLA-4, The Journal of Immunology 5549

FIGURE 7. CTLA-4 regulates immunological synapse organization in Th2 and Th1 cells. A, Th2 cells from CTLA-4Ϫ/Ϫ AND transgenic mice were transfected with virus containing either an empty GFP vector or a CTLA-4 GFP vector and stained intracellularly for CTLA-4. CTLA-4 expression levels on GFPϩ cells were analyzed. B, GFPϩ CTLA- 4Ϫ/Ϫ Th2 cells were sorted and stimulated with peptide-pulsed B cells. C, Quantitation of TCR clustering at the Th2 cell/B cell site in the presence or Downloaded from absence of CTLA-4 (as described in Fig. 1C). Dif- ference in TCR-clustering frequency between Th2 cells transfected with CTLA-4 vs empty vector is signiﬁcant with p Ͻ 0.0001. This experiment was repeated three times. D, CTLA-4-intact Th1 cells were transfected and stained as in A. E, GFPϩ wild- http://www.jimmunol.org/ type Th1 cells were sorted and stimulated with peptide-pulsed B cells. F, Quantitation of TCR clustering at the Th1 cell/B cell site in the presence or absence of CTLA-4 (as described in Fig. 1C). Dif- ference in TCR-clustering frequency between Th1 cells transfected with CTLA-4 vs empty vector is signiﬁcant with p Ͻ 0.0001. This experiment was repeated three times. by guest on September 24, 2021

CD80, and CD86 triple-knockout mice are healthy (45). CTLA-4, duction of CTLA-4 was confirmed by flow cytometry (Fig. 7A). CD80, and CD86 triple-knockout mice were bred onto AND trans- Th2 cells given the control virus (GFP alone) still clustered the genic Rag2 knockout mice. Effective Th1/Th2 generation was con- TCR at a frequency consistent with previous experiments (Fig. 7, firmed using intracellular cytokine staining for IL-4 and IFN-␥ B and C). In contrast, Th2 cells in which CTLA-4 was restored lost (data not shown). Th1 and Th2 cells from CTLA-4-deficient and the ability to cluster the TCR (Fig. 7, B and C). Therefore, CTLA-4 CTLA-4 heterozygous littermates were restimulated with resting B expression is both necessary and sufficient for the observed block cells and then examined for TCR clustering. In the cells generated in Th2 TCR clustering. from the CTLA-4 heterozygous littermates, Th1 cells clustered the To investigate whether CTLA-4 could block the clustering of TCR at the site of T cell/B cell conjugation, but Th2 cells did not the TCR in Th1 cells, the same CTLA-4 construct was introduced (Fig. 6, A, upper two rows, and B). In contrast, both Th1 and Th2 into wild-type Th1 cells. GFPϩ cells were sorted, stimulated with cells generated from the CTLA-4-deficient mice clustered the TCR peptide-pulsed resting B cells, and examined for TCR clustering. with equal frequency (Fig. 6, A, lower two rows, and B). Interestingly, GFPϩ cells had only a very small increase in To confirm the role of CTLA-4 in Th2 synapse organization, CTLA-4 expression (Fig. 7D), but even with this small increase in CTLA-4 was reintroduced into Th2 cells generated from CTLA- CTLA-4 expression, there was a corresponding reduction in the 4-deficient mice. A retroviral construct was prepared containing frequency of TCR clustering when stimulated with resting B cells wild-type murine CTLA-4 coexpressed with GFP. Th2 cells were (Fig. 7, E and F). generated from CTLA-4-deficient TCR transgenic mice, trans- To better understand the quantity of CTLA-4 that is required to duced with the viral construct during restimulation, and allowed to block TCR clustering at the synapse site, we compared levels of rest. GFPϩ cells were sorted, stimulated with peptide-pulsed rest- expression of CTLA-4 under conditions in which the TCR did, or ing B cells, and examined for TCR clustering. Successful reintro- did not cluster. We used the viral transduction experiments 5550 CTLA-4 AND THE CD4 T CELL IMMUNOLOGICAL SYNAPSE because they provide matched synapse and CTLA-4 expression The difference we observed in CTLA-4 levels between Th1 and data, and included both the Th2 and Th1 experiments. To allow for Th2 cells may play additional roles in the differential regulation of comparison between experiments, we normalized the mean fluo- signaling and activation between these two effector subsets. We rescent intensity of CTLA-4 against the matched isotype control have looked at cytokine production by wild-type Th2 cells stim- stain in each experiment, providing a value representing relative ulated with resting vs activated B cells, and also compared cyto- expression of CTLA-4. We then averaged the values for conditions kine production between CTLA-4-deficient or -intact Th2 cells, to in which the TCR did, or did not cluster at the synapse site, which see whether there is a correlation between synapse organization were 1.5 and 4.5, respectively. This represents a 3-fold increase, and cytokine production. As expected, we found that increased which is consistent with our observed differences in CTLA-4 ex- IL-2 production correlated with increased CD28 signals, as has pression between Th1 and Th2 cells at both the protein and RNA been previously published (46), but saw no additional differences levels (data not shown). (data not shown). In addition to differences in cytokine production, Th2 cells have been shown to have defects in tyrosine phosphor- Discussion ylation of Fyn and ZAP70, in sustained calcium mobilization, and In this study, we have demonstrated that whereas both Th1 and in ability to respond to low-affinity peptide/MHC (13, 47–50). Th2 cells cluster the TCR at the site of contact with activated Some of these differences could be the result of differences in APCs, only Th1 cells are able to do so with resting B cells. Al- relative levels of CTLA-4 in effector Th2 vs effector Th1 or naive though both Th1 and Th2 cells require CD80/CD86 for optimal CD4 T cells. We have shown that the higher levels of CTLA-4 in TCR clustering, the difference in synapse organization can be ex- resting Th2 cells block the clustering of the TCR at the B cell plained by the higher expression level of CTLA-4 in Th2 cells, conjugation site. This lack of clustering could lead to a reduction Downloaded from including expression at the cell surface. This increased expression in avidity that would block full activation of the Th2 cell when the of CTLA-4 inhibits TCR clustering. This suggests a model in TCR affinity is low. Th2 differentiation has been associated with which low-level expression of CD80/CD86 on resting B cells is lower affinity TCR interactions (13, 51, 52). If the normal Th2 sufficient to provide the activating signals via CD28 on Th1 cells TCR repertoire is of lower specificity, then Th2 cells might only be needed for aggregation of the TCR. Higher levels of CTLA-4 ex- able to be fully activated by APCs that have received innate acti-

pression on Th2 cells, however, lead to inhibitory signals when vation signals leading to increased expression of CD80/CD86 that http://www.jimmunol.org/ CD80/CD86 expression levels on the APC are low, resulting in a enable them to trigger clustering of the TCR. In this way, CTLA-4 loss of TCR clustering. When the expression of CD80/CD86 is may act as an additional level of control over Th2 responses. Our increased, as is observed on activated B cells or DCs, dominant work suggests that the APC may influence other Th2 signaling CD28 signaling is restored, leading to clustering of TCR events, and that the differences in signaling observed between Th1 molecules. and Th2 might be overcome when different APCs are used to We have demonstrated in this study a new role for CTLA-4 as stimulate. the critical regulator of Th2 immunological synapse organization Our findings provide an interesting mechanism of control and are the first to demonstrate a differential role for CTLA-4 in unique to Th2 cell/B cell interactions, in which a B cell must first different subsets of primary T cells. In contrast to wild-type Th2 receive an activation signal itself before it is able to signal the by guest on September 24, 2021 cells, which fail to cluster the TCR when stimulated with resting B clustering of the TCR at the site of contact with its cognate Th2 cells, TCR clustering occurs in Th2 cells generated from CTLA- cell. This extra activation step is not required in Th1 cells, sug- 4-deficient mice. The subsequent loss of TCR clustering when gesting a unique function for immunological synapse organization CTLA-4 is reintroduced into these cells demonstrates that this is a in Th2/B cell interactions, and possibly a need for tighter control function of CTLA-4. We have also shown that APC types vary in via innate immune signals. their ability to cluster the TCR at the site of conjugation with Th2 cells, which may have important implications in vivo. This is crit- Acknowledgments ical, because many studies on immunological synapse formation We thank P. Ranney for animal support; J. Czyzyk for assistance with have used cell lines as APCs, or other artificial stimuli, which may retroviral work; A. Sharpe for the CTLA-4Ϫ/Ϫ/CD80Ϫ/Ϫ/CD86Ϫ/Ϫ mice; fail to detect these differences. For example, in some of the early K. Murphy for the retroviral vector; G. Nolan for the Phoenix Ecotropic synapse studies, B cell lymphomas were used to stimulate D10 packaging cells; S. Anderson for helpful discussion; and S. Connolly, cells (a Th2 clone) (1). Given our data with activated B cells, it is H. Chen, and J. Czyzyk for discussion of manuscript. not surprising that a B cell lymphoma would yield similar results, masking the differences we have observed between Th1 and Th2 Disclosures cells. The authors have no financial conflict of interest. This model fits with what we know about the differing abilities of CTLA-4 and CD28 to bind to their shared ligands, CD80 and References CD86. CTLA-4 is able to bind to both CD80 and CD86 with 1. Monks, C. R., B. A. Freiberg, H. Kupfer, N. Sciaky, and A. Kupfer. 1998. Three- dimensional segregation of supramolecular activation clusters in T cells. Nature higher affinities and avidities than CD28. Binding studies have 395: 82–86. found that CTLA-4 has ϳ20-fold higher affinity for CD80 and 2. Lee, K. H., A. D. Holdorf, M. L. Dustin, A. C. Chan, P. M. Allen, and A. S. Shaw. ϳ 2002. T cell receptor signaling precedes immunological synapse formation. Sci- 8-fold higher affinity for CD86 when compared with CD28 (23). ence 295: 1539–1542. In addition, both structural and binding studies have shown that 3. Anton van der Merwe, P., S. J. Davis, A. S. Shaw, and M. L. Dustin. 2000. whereas CD28 forms monovalent dimers, CTLA-4 forms bivalent Cytoskeletal polarization and redistribution of cell-surface molecules during T cell antigen recognition. Semin. Immunol. 12: 5–21. dimers that enable it to form highly stable lattice structures with 4. Wulfing, C., M. D. Sjaastad, and M. M. Davis. 1998. Visualizing the dynamics CD80 and CD86, resulting in an increase in its avidity for these of T cell activation: intracellular adhesion molecule 1 migrates rapidly to the T molecules by ϳ100-fold (21–23). The combination of this in- cell/B cell interface and acts to sustain calcium levels. Proc. Natl. Acad. Sci. USA 95: 6302–6307. creased avidity and affinity gives CTLA-4 a binding advantage of 5. Xavier, R., T. Brennan, Q. Li, C. McCormack, and B. Seed. 1998. Membrane ϳ3 orders of magnitude over CD28. This accounts for our obser- compartmentation is required for efficient T cell activation. Immunity 8: 723–732. 6. Rodgers, W., and J. K. Rose. 1996. Exclusion of CD45 inhibits activity of p56lck vation at limiting CD80/CD86 molecule expression where associated with glycolipid-enriched membrane domains. J. Cell Biol. 135: CTLA-4 inhibition of clustering occurs. 1515–1523. The Journal of Immunology 5551

7. Krummel, M. F., M. D. Sjaastad, C. Wulfing, and M. M. Davis. 2000. Differential 31. Eisenbarth, S. C., D. A. Piggott, J. W. Huleatt, I. Visintin, C. A. Herrick, and clustering of CD4 and CD3␨ during T cell recognition. Science 289: 1349–1352. K. Bottomly. 2002. Lipopolysaccharide-enhanced, Toll-like receptor 4-dependent 8. Villalba, M., N. Coudronniere, M. Deckert, E. Teixeiro, P. Mas, and A. Altman. T helper cell type 2 responses to inhaled antigen. J. Exp. Med. 196: 1645–1651. 2000. A novel functional interaction between Vav and PKC␪ is required for 32. Tremuth, L., S. Kreis, C. Melchior, J. Hoebeke, P. Ronde, S. Plancon, K. Takeda, TCR-induced T cell activation. Immunity 12: 151–160. and N. Kieffer. 2004. A fluorescence cell biology approach to map the second 9. Hailman, E., W. R. Burack, A. S. Shaw, M. L. Dustin, and P. M. Allen. 2002. integrin-binding site of talin to a 130-amino acid sequence within the rod domain. ϩ ϩ Immature CD4 CD8 thymocytes form a multifocal immunological synapse J. Biol. Chem. 279: 22258–22266. with sustained tyrosine phosphorylation. Immunity 16: 839–848. ϩ ϩ 33. Nayal, A., D. J. Webb, and A. F. Horwitz. 2004. Talin: an emerging focal point 10. Ebert, P. J., J. F. Baker, and J. A. Punt. 2000. Immature CD4 CD8 thymocytes of adhesion dynamics. Curr. Opin. Cell Biol. 16: 94–98. do not polarize lipid rafts in response to TCR-mediated signals. J. Immunol. 165: 34. Calderwood, D. A., R. Zent, R. Grant, D. J. Rees, R. O. Hynes, and 5435–5442. M. H. Ginsberg. 1999. The talin head domain binds to integrin ␤ subunit cyto- 11. O’Keefe, J. P., K. Blaine, M. L. Alegre, and T. F. Gajewski. 2004. Formation of plasmic tails and regulates integrin activation. J. Biol. Chem. 274: 28071–28074. a central supramolecular activation cluster is not required for activation of naive ϩ 35. Calderwood, D. A., and M. H. Ginsberg. 2003. Talin forges the links between CD8 T cells. Proc. Natl. Acad. Sci. USA 101: 9351–9356. integrins and actin. Nat. Cell Biol. 5: 694–697. 12. Purbhoo, M. A., D. J. Irvine, J. B. Huppa, and M. M. Davis. 2004. T cell killing 36. Meyer zum Bueschenfelde, C. O., J. Unternaehrer, I. Mellman, and K. Bottomly. does not require the formation of a stable mature immunological synapse. Nat. 2004. Regulated recruitment of MHC class II and costimulatory molecules to Immunol. 5: 524–530. lipid rafts in dendritic cells. J. Immunol. 173: 6119–6124. 13. Balamuth, F., D. Leitenberg, J. Unternaehrer, I. Mellman, and K. Bottomly. 2001. Distinct patterns of membrane microdomain partitioning in Th1 and Th2 cells. 37. Dustin, M. L., and A. S. Shaw. 1999. Costimulation: building an immunological Science Immunity 15: 729–738. synapse. 283: 649–650. 14. Balamuth, F., J. L. Brogdon, and K. Bottomly. 2004. CD4 raft association and 38. Alegre, M. L., P. J. Noel, B. J. Eisfelder, E. Chuang, M. R. Clark, S. L. Reiner, signaling regulate molecular clustering at the immunological synapse site. J. Im- and C. B. Thompson. 1996. Regulation of surface and intracellular expression of munol. 172: 5887–5892. CTLA4 on mouse T cells. J. Immunol. 157: 4762–4770. 15. Ward, S. G. 1996. CD28: a signalling perspective. Biochem. J. 318: 361–377. 39. Iida, T., H. Ohno, C. Nakaseko, M. Sakuma, M. Takeda-Ezaki, H. Arase, 16. Bromley, S. K., A. Iaboni, S. J. Davis, A. Whitty, J. M. Green, A. S. Shaw, E. Kominami, T. Fujisawa, and T. Saito. 2000. Regulation of cell surface expression of CTLA-4 by secretion of CTLA-4-containing lysosomes upon activa- Downloaded from A. Weiss, and M. L. Dustin. 2001. The immunological synapse and CD28-CD80 ϩ interactions. Nat. Immunol. 2: 1159–1166. tion of CD4 T cells. J. Immunol. 165: 5062–5068. 17. Wulfing, C., and M. M. Davis. 1998. A receptor/cytoskeletal movement triggered 40. Leung, H. T., J. Bradshaw, J. S. Cleaveland, and P. S. Linsley. 1995. Cytotoxic by costimulation during T cell activation. Science 282: 2266–2269. T lymphocyte-associated molecule-4, a high-avidity receptor for CD80 and 18. Viola, A., S. Schroeder, Y. Sakakibara, and A. Lanzavecchia. 1999. T lympho- CD86, contains an intracellular localization motif in its cytoplasmic tail. J. Biol. cyte costimulation mediated by reorganization of membrane microdomains. Sci- Chem. 270: 25107–25114. ence 283: 680–682. 41. Linsley, P. S., J. Bradshaw, J. Greene, R. Peach, K. L. Bennett, and R. S. Mittler. 19. Darlington, P. J., M. L. Baroja, T. A. Chau, E. Siu, V. Ling, B. M. Carreno, and 1996. Intracellular trafficking of CTLA-4 and focal localization towards sites of

J. Madrenas. 2002. Surface cytotoxic T lymphocyte-associated antigen 4 parti- TCR engagement. Immunity 4: 535–543. http://www.jimmunol.org/ tions within lipid rafts and relocates to the immunological synapse under condi- 42. Valk, E., R. Leung, H. Kang, K. Kaneko, C. E. Rudd, and H. Schneider. 2006. T tions of inhibition of T cell activation. J. Exp. Med. 195: 1337–1347. cell receptor-interacting molecule acts as a chaperone to modulate surface ex- 20. Egen, J. G., and J. P. Allison. 2002. Cytotoxic T lymphocyte antigen-4 accumu- pression of the CTLA-4 coreceptor. Immunity 25: 807–821. lation in the immunological synapse is regulated by TCR signal strength. Immu- 43. Waterhouse, P., J. M. Penninger, E. Timms, A. Wakeham, A. Shahinian, nity 16: 23–35. K. P. Lee, C. B. Thompson, H. Griesser, and T. W. Mak. 1995. Lymphoprolif- 21. Schwartz, J. C., X. Zhang, A. A. Fedorov, S. G. Nathenson, and S. C. Almo. erative disorders with early lethality in mice deficient in CTLA-4. Science 270: 2001. Structural basis for co-stimulation by the human CTLA-4/B7-2 complex. 985–988. Nature 410: 604–608. 44. Tivol, E. A., F. Borriello, A. N. Schweitzer, W. P. Lynch, J. A. Bluestone, and 22. Stamper, C. C., Y. Zhang, J. F. Tobin, D. V. Erbe, S. Ikemizu, S. J. Davis, A. H. Sharpe. 1995. Loss of CTLA-4 leads to massive lymphoproliferation and M. L. Stahl, J. Seehra, W. S. Somers, and L. Mosyak. 2001. Crystal structure of fatal multiorgan tissue destruction, revealing a critical negative regulatory role of the B7-1/CTLA-4 complex that inhibits human immune responses. Nature 410: CTLA-4. Immunity 3: 541–547. 608–611. 45. Mandelbrot, D. A., A. J. McAdam, and A. H. Sharpe. 1999. B7-1 or B7-2 is by guest on September 24, 2021 23. Collins, A. V., D. W. Brodie, R. J. Gilbert, A. Iaboni, R. Manso-Sancho, required to produce the lymphoproliferative phenotype in mice lacking cytotoxic B. Walse, D. I. Stuart, P. A. van der Merwe, and S. J. Davis. 2002. The interaction T lymphocyte-associated antigen 4 (CTLA-4). J. Exp. Med. 189: 435–440. properties of costimulatory molecules revisited. Immunity 17: 201–210. 46. Lindstein, T., C. H. June, J. A. Ledbetter, G. Stella, and C. B. Thompson. 1989. 24. Kaye, J., M. L. Hsu, M. E. Sauron, S. C. Jameson, N. R. Gascoigne, and ϩ Regulation of lymphokine messenger RNA stability by a surface-mediated T cell S. M. Hedrick. 1989. Selective development of CD4 T cells in transgenic mice activation pathway. Science 244: 339–343. expressing a class II MHC-restricted antigen receptor. Nature 341: 746–749. 47. Sloan-Lancaster, J., T. H. Steinberg, and P. M. Allen. 1997. Selective loss of the 25. Levin, D., S. Constant, T. Pasqualini, R. Flavell, and K. Bottomly. 1993. Role of ϩ calcium ion signaling pathway in T cells maturing toward a T helper 2 phenotype. dendritic cells in the priming of CD4 T lymphocytes to peptide antigen in vivo. J. Immunol. 159: 1160–1168. J. Immunol. 151: 6742–6750. 26. Lutz, M. B., N. Kukutsch, A. L. Ogilvie, S. Rossner, F. Koch, N. Romani, and 48. Gajewski, T. F., S. R. Schell, and F. W. Fitch. 1990. Evidence implicating uti- G. Schuler. 1999. An advanced culture method for generating large quantities of lization of different T cell receptor-associated signaling pathways by TH1 and highly pure dendritic cells from mouse bone marrow. J. Immunol. Methods 223: TH2 clones. J. Immunol. 144: 4110–4120. 2ϩ 77–92. 49. Fanger, C. M., A. L. Neben, and M. D. Cahalan. 2000. Differential Ca influx, 2ϩ 27. Leitenberg, D., Y. Boutin, S. Constant, and K. Bottomly. 1998. CD4 regulation KCa channel activity, and Ca clearance distinguish Th1 and Th2 lymphocytes. of TCR signaling and T cell differentiation following stimulation with peptides of J. Immunol. 164: 1153–1160. different affinities for the TCR. J. Immunol. 161: 1194–1203. 50. Tamura, T., H. Nakano, H. Nagase, T. Morokata, O. Igarashi, Y. Oshimi, 28. Pierre, P., L. K. Denzin, C. Hammond, J. R. Drake, S. Amigorena, P. Cresswell, S. Miyazaki, and H. Nariuchi. 1995. Early activation signal transduction path- and I. Mellman. 1996. HLA-DM is localized to conventional and unconventional ways of Th1 and Th2 cell clones stimulated with anti-CD3: roles of protein MHC class II-containing endocytic compartments. Immunity 4: 229–239. tyrosine kinases in the signal for IL-2 and IL-4 production. J. Immunol. 155: 29. Piggott, D. A., S. C. Eisenbarth, L. Xu, S. L. Constant, J. W. Huleatt, 4692–4701. C. A. Herrick, and K. Bottomly. 2005. MyD88-dependent induction of allergic 51. Boyton, R. J., and D. M. Altmann. 2002. Is selection for TCR affinity a factor in Th2 responses to intranasal antigen. J. Clin. Invest. 115: 459–467. cytokine polarization? Trends Immunol. 23: 526–529. 30. Herrick, C. A., H. MacLeod, E. Glusac, R. E. Tigelaar, and K. Bottomly. 2000. 52. Tao, X., S. Constant, P. Jorritsma, and K. Bottomly. 1997. Strength of TCR signal Th2 responses induced by epicutaneous or inhalational protein exposure are dif- determines the costimulatory requirements for Th1 and Th2 CD4ϩ T cell differ- ferentially dependent on IL-4. J. Clin. Invest. 105: 765–775. entiation. J. Immunol. 159: 5956–5963.