This information is current as of December 22, 2018. The TCR ζ-chain Immunoreceptor Tyrosine-Based Activation Motifs Are Sufficient for the Activation and Differentiation of Primary T Lymphocytes Terrence L. Geiger, David Leitenberg and Richard A. Flavell J Immunol 1999; 162:5931-5939; ; http://www.jimmunol.org/content/162/10/5931 References Subscription Permissions Email Alerts This article cites 47 articles, 19 of which you can access for free at: http://www.jimmunol.org/content/162/10/5931.full#ref-list-1 Why The JI? Submit online. Rapid Reviews! 30 days* from submission to initial decision No Triage! Every submission reviewed by practicing scientists Fast Publication! 4 weeks from acceptance to publication *average Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Submit copyright permission requests at: http://www.aai.org/about/publications/ji/copyright.html Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts Downloaded from http://www.jimmunol.org/ by guest on December 22, 2018 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 1999 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606.
The TCR -Chain Immunoreceptor Tyrosine-Based Activation Motifs Are Sufficient for the Activation and Differentiation of Primary T Lymphocytes 1 Terrence L. Geiger,* David Leitenberg,* and Richard A. Flavell 2 * The TCR complex signals through a set of 10 intracytoplasmic motifs, termed immunoreceptor tyrosine-based activation motifs (ITAMs), contained within the -, -, -, and -chains. The need for this number of ITAMs is uncertain. Limited and contradictory studies have examined the ability of subsets of the TCR s ITAMs to signal into postthymic primary T lymphocytes. To study signaling by a restricted set of ITAMs, we expressed in transgenic mice a chimeric construct containing the IA s class II MHC extracellular and transmembrane domains linked to the cytoplasmic domain of the TCR -chain. Tyrosine phosphorylation and receptor cocapping studies indicate that this chimeric receptor signals T cells independently of the remainder of the TCR. We show that CD4 and CD8 primary T cells, as well as naive and memory T cells, are fully responsive to stimulation through the IA s - receptor. Further, IA s - stimulation can induce primary T cell differentiation into CTL, Th1, and Th2 type cells. These results show that the -chain ITAMs, in the absence of the,, and ITAMs, are sufficient for the activation and functional maturation of primary T lymphocytes. It also supports the isolated use of the -chain ITAMs in the development of surrogate TCRs for therapeutic purposes. The Journal of Immunology, 1999, 162: 5931 5939. The TCR is a multimeric complex that includes,,,,, and polypeptides. These associate through noncovalent interactions (1, 2). Whereas the - and -chains recognize peptide Ags presented by MHC, signaling is exclusively mediated by the -, -, -, and -chains (3, 4). These latter TCR components include well-conserved tyrosine-containing domains, immunoreceptor tyrosine-based activation motifs (ITAMs) 3, that are rapidly phosphorylated following TCR engagement (5, 6). The ITAM tyrosines are embedded within a Y-XX-L-X 7 8 -Y-XX-L consensus sequence. Phosphorylation of the tyrosines permits TCR association with Src homology 2 (SH2) domain-containing proteins. These, in turn, couple initial TCR phosphorylation with downstream signaling events. The large number of signaling domains within the TCR is unusual among cellular receptors. With a typical structural composition of 2 zeta 2, a total of 10 ITAM domains are present within the TCR s chains. Three of these are found within each -chain, which form a disulfide-linked homodimer (7). A single ITAM is additionally contained within each of the -, -, and -chains. The reason for this abundance of signaling domains is *Section of Immunobiology, and Department of Laboratory Medicine, Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, CT 06510 Received for publication September 17, 1998. Accepted for publication February 18, 1999. 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 with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by National Institutes of Health Grants AI01480 (to T.L.G.) and AI01314 (to D.L.), and by the Howard Hughes Medical Institute. R.A.F. is an investigator of the Howard Hughes Medical Institute. D.L. is supported in part by an Arthritis Investigator Award. 2 Address correspondence and reprint requests to Dr. Richard A. Flavell, Section of Immunobiology/Howard Hughes Medical Institute, Yale University School of Medicine, 310 Cedar Street, FMB 412, New Haven, CT 06520. E-mail address: richard.flavell@qm.yale.edu 3 Abbreviations used in this paper: ITAM, immunoreceptor tyrosine-based activation motif; SH2, Src homology 2; MCS, multiple cloning site; CHO, Chinese hamster ovary; LCR, locus control region. unclear. One hypothesis is that different ITAMs couple with distinct signal-transducing molecules, thus providing qualitatively different signals to the T cell. Indeed, there are few conserved amino acids within the ITAM motif, and it may be expected that neighboring residues influence the fine specificity of SH2 protein binding. This idea is conceptually appealing because it can explain how proximal signaling events may induce different functional outcomes within a T cell. The type of signal transmitted by a TCR, e.g., agonist, partial agonist, or antagonist, could thus be determined by the specific set of ITAMs that are phosphorylated (8). Some studies support the idea that different ITAMs have qualitatively distinct signaling tasks. For example, ITAM peptides show differential binding to several SH2-containing proteins, including Syk, Lyn, Shc, Grb2, and Plc- 1 (9). Of the -chain ITAMs, cytoskeletal actin preferentially associates with the C-terminal ITAM in T cell hybridomas (10). Further, functional studies using BW5147 T cell hybridomas transfected with various ITAMdeficient TCRs or or chimeric receptors show differential Ca mobilization after stimulation (11). A second hypothesis for the abundance of ITAMs relates to the quantitative need for signaling. The TCR is of relatively low abundance and forms low-affinity interactions with its MHC-peptide ligand. Multiple ITAMs may be necessary to amplify the signal of a sparse number of effective engagements with ligand. Some evidence supports this hypothesis as well. Mice made genetically deficient for the TCR -chain (and hence 6 of the 10 ITAMs) and then transgenically reconstituted with -chains containing 0, 1, or 3 ITAMs display normal thymic maturation (12, 13). Positive and negative selection proceed normally, yet the strength of selection is related to the number of ITAM domains. T lymphocytes capable of responding to mitogen or anti-cd3-mediated stimulation populate the peripheral lymphoid organs. In addition, phosphorylation patterns of known signal-transducing proteins in stimulated thymocytes from these mice are essentially identical, regardless of the number of ITAMs (14). Indeed, as further support for a quantitative role for ITAMs in signaling, T cell hybridomas can be activated through even a single ITAM domain (15, 16). Copyright 1999 by The American Association of Immunologists 0022-1767/99/$02.00
5932 -CHAIN-MEDIATED PRIMARY T CELL ACTIVATION Understanding the role of individual or sets of ITAMs in T cell signal transduction is not only important with respect to the basic mechanics of TCR signaling, it is also assuming increasing clinical relevance. A number of in vitro and in vivo studies have demonstrated that chimeric receptors made with or other TCR ITAMs linked to extracellular recognition domains, such as Fv fragments, can transduce signals into T cell clones or hybridomas (17 23). This may result in proliferation, cytokine production, or induction of cytolytic activity. Indeed, initial clinical trials are being conducted with chimeric receptors to assess their potential in the therapy of ovarian carcinoma (23). Effective use of such surrogate TCRs will require an understanding of the signaling capabilities of receptors containing limited numbers of ITAMs in primary T lymphocytes. Yet, few data are published on the capacity of limited sets of ITAMs to signal independently of the TCR in postthymic T cells. One study, analyzing a Fv- chimera expressed on T cells in transgenic mice, failed to demonstrate effective signaling into primary T lymphocytes unless they were first preactivated through their native TCR (24). The authors concluded that -mediated signals are insufficient to prime resting T cells. In another study, the IL-2 receptor (Tac) extracellular and transmembrane domain was linked to (Tac- ) or (Tac- ) and transgenically expressed on T cells (25). Stimulation through these receptors in the presence of Con A supernatant induced a proliferative response. Signaling through Tac- or Tac- chimeric receptors into thymocytes also appeared normal. These authors concluded that different CD3 components deliver qualitatively similar signals. The contradictory nature of these studies leaves doubt as to the signal transduced by the -chain ITAMs into primary T cells. To address this issue, we have expressed a chimeric molecule containing class II MHC extracellular and transmembrane domains and the cytoplasmic domain in transgenic mice. We find that this chimeric receptor signals T cells independently of the remaining components of the TCR. Further, the IA s - receptor is capable of activating both memory and naive T cells, as well as promoting T cell differentiation into CTL, Th1, and Th2 T cells. These results demonstrate that the subset of ITAMs included within the -chain ITAMs, independent of the,, and ITAMs, are sufficient for primary T cell activation and functional maturation. Materials and Methods Generation of chimeric constructs For the IA s - chimeric construct, the 89 101 peptide epitope of myelin basic protein was genetically linked to the 5 end by modifying the previously described A b with E d peptide construct ptz18r (26, 27). Briefly, oligonucleotides (primers A and B, below) were used to PCR isolate the SacI-EcoRI 5 fragment of the construct, which was subcloned into a multiple cloning site (MCS)-modified pbs vector (Stratagene, La Jolla, CA). The BamHI-NheI fragment was replaced with synthetic annealed 5 and 3 oligonucleotides containing the 89 101 peptide fragment of myelin basic protein. The mid portion of this construct containing the extracellular and transmembrane portions of IA s was produced from first strand cdna generated from SJL/J spleen, as described (28). Oligonucleotide primers (primers C and D) were used to insert 5 SacI and 3 XbaI restriction sites by PCR, and this fragment was subcloned into a MCS-modified pbs. The cytoplasmic portion of the -chain was isolated by PCR (primers E and F) using the psr CD3 construct with a 5 XbaI and 3 XhoI site being created and subcloned into pbs. These three fragments of the final construct were sequenced, ligated together, and then placed in either the phcd2-va (29, 30) or plxsn expression vectors (31). The class II portion of the IA s - construct had 5 EcoRI and 3 XbaI sites inserted by PCR (primers G and H) using IA s cdna ph APR-1-neo (gift of H. McDevitt, Stanford University, Palo Alto, CA). This was subcloned into a MCS-modified pbs, sequenced, linked to the cytoplasmic -chain, and subcloned into the prv-hyg (32) or phcd2-va expression vectors. Oligonucleotides were synthesized at the Keck Biotechnology Resource Laboratory at Yale University. Sequences were: primer A: 5 -ATTCGAAGATCTGAAT TCTTAGAGATGGC-3 ; primer B: 5 -GAAATGCCTTTCAGAGCTCCCACC TCC-3 ; primer C: 5 -GGGCGGGAGCTCCGAAAGGCATTTCGTGTTCC-3 ; primer D: 5 -GCCTCGCCCGGGTTTCTGACTCCTGTGACGGAT-3 ; primer E: 5 -GCCCTGCCCGGGAGAGCAAAATTCAGCAGGA-3 ; primer F:5 -GTACCACTCGAGATTTAGTTAGGAAGAGCA-3 ; primer G: 5 -AG GTCGAATTCGCAGAGACCTCCCAGAGACCAGGATGC-3 ; and primer H: 5 -GGAGGTCCCGGGTGATCGCAGGCCTTGAATGATGAA-3. Transgenic mouse production Insert DNA, from the IA s - and IA s - constructs subcloned into phcd2- VA, was removed from plasmid sequence by cleavage with KpnI and NotI and isolated by gel electrophoresis and electrolelution. The inserts were further purified, mixed in equimolar quantities, and coinjected into (B6xC3H)F 2 day 1 embryos. Progeny were analyzed by restriction fragment length polymorphism and Southern blot analysis using probes specific for the IA s and IA s genes. Mice were subsequently bred with BALB/c or SJL/J mice and analyzed by flow cytometry of peripheral blood stained with class II MHC and CD8-specific Abs. All procedures involving animals were performed after review and approval by the Yale Animal Care and Use Committee. Abs and flow cytometry Monoclonal FITC-, biotin-, PE-, Texas Red-, or CyChrome-conjugated CD4, CD8, CD44, CD69, CD25, CD45RB, streptavidin, and avidin were purchased from PharMingen (San Diego, CA), IA.B2 from Devaron (Dayton, NJ) or Biodesign International (Kennebunk, ME), and 4G10 from Upstate Biotechnology (Lake Placid, NY). Polyclonal rabbit anti-tcr- was produced as described (33). Monoclonal anti-il-5 and IFN- Abs were purchased from PharMingen and used following manufacturer s instructions. 11B11 (anti-il-4), XMG1.2 (anti-ifn- ), GK1.5 (anti-cd4), 53-6.7 (anti-cd8), HB-191 (anti-nk) and 2C11 (anti-cd3 ) were purified form hybridoma supernatants as described (34) and spectrophotometrically quantitated. Flow cytometry was performed on a FACScalibur and sterile cell sorting on a FACStar Plus (Becton Dickinson, Mountain View, CA) using Cellquest software. Calcium mobilization Calcium signaling following Ab cross-linking was monitored as described previously (35). Briefly, T cells loaded with 5 M fluo-3/am ester (Molecular Probes, Eugene, OR) were plated by centrifugation in 96-well plates at a concentration of 5 10 5 cells/100 l. The cells were then scanned using the ACAS 570 video laser cytometer (Meridian Instruments, Okemos, IL). The cells were preincubated for 5 min with saturating amounts of the following mabs: 2C11 (anti-cd3 ), GK1.5 (anti-cd4), and IA.B2 (anti-class II MHC). After initiation of scanning, the Abs were cross-linked with goat anti-rat IgG (ICN Pharmaceuticals, Costa Mesa, CA). The initial average fluorescence of each cell was digitized and normalized to 1, and the results are expressed as changes in normalized fluorescence intensity of individual cells over time. The percentage of responding cells was determined by dividing the number of cells demonstrating an increase in intracellular calcium of 50% by the total number of scanned cells. Protein biochemistry and immunoprecipitation T cells were stimulated by Ab cross-linking, as described previously, for 2 min and lysed in ice-cold lysis buffer (20 mm Tris (ph 7.2), 1% Nonidet P-40, 150 mm NaCl, 1 mm MgCl 2, 1 mm EGTA) containing protease and phosphatase inhibitors (10 mm Na 4 P 2 O 7 10H 2 O,1mMNa 3 VO 4,50mM NaF, 1 mm PMSF, 10 g/ml leupeptin, and 10 g/ml aprotinin), and nuclear material was removed as previously described (36). Cell lysates were incubated for 1.5 h with protein A-Sepharose CL-4B beads that had been pretreated with anti- -chain rabbit antisera prepared in our laboratory. Immunoprecipitates were washed twice in 1% Nonidet P-40 and twice in 0.1% Nonidet P-40 lysis buffer. Phosphotyrosine-containing proteins were detected following 12% SDS-PAGE and blotting with antiphosphotyrosine mab 4G10 followed by goat anti-mouse IgG HRP conjugate (Bio-Rad, Hercules, CA) and detected by enhanced chemiluminescence, as described by the manufacturer (Amersham, Buckinghamshire, England). -chain was detected by blotting with rabbit anti- antisera and Protein A-conjugated HRP (Sigma, St. Louis, MO). Receptor capping analysis Receptor cocapping studies were performed as described (37). Briefly, 5 10 5 lymph node cells were incubated in Bruff medium/10% FCS with unconjugated primary Ab (IA.B2 or 2C11) at 37 C for 30 min. The cells were washed and then capped with FITC-conjugated goat anti-mouse
The Journal of Immunology 5933 IgG at 37 C for 30 min. After washing three times, the cells were fixed with 0.5% paraformaldehyde, washed a further three times, and analyzed or stained at 4 C with biotinylated 2C11 or Y-3 (anti-class II) Ab in the presence of 0.1% sodium azide, followed by avidin-texas Red staining. Alternatively, biotinylated Y-3 or 2C11 were capped with avidin-texas Red, and follow-up staining was performed with FITC-conjugated 2C11 or IA.B2 Abs. Cell suspensions were analyzed with a Zeiss Axiophot fluorescent microscope (Thornwood, NY). Cytotoxicity assays Total splenocytes, stimulated 5 days with plate-bound 2C11 or IA.B2 Abs, were used as effectors. Chinese hamster ovary (CHO) or Fc RIIB1-transfected CHO (CHO-B1) cells were used as targets (gift of I. Mellman, Yale University, New Haven, CT). Lysis was directed by the addition of 0.5 5 g/ml 2C11 to the incubation medium during the assay. A 4.5-h 51 Crrelease assay was performed as described (38). T cell stimulation and naive T cell or T cell subset isolation A total of 5 10 4 or 10 5 splenocyte responders was stimulated per 96-well plate well coated with the designated type and 5 g/ml or the listed concentration of Ab. After 2 3 days, the cells were pulsed for 16 20 h with 1 Ci [ 3 H]thymidine, then harvested onto filtermats. Proliferation was determined by [ 3 H]thymidine incorporation measured by liquid scintillation counting. All samples were analyzed in duplicate. T cell subsets were enriched by magnetic bead depletion per the manufacturer s instructions (Perseptive Biosystems, Framingham, MA). CD8 or CD4 subsets were obtained from mixed lymph node cells depleted by incubating with monoclonal rat anti-cd4 or anti-cd8 Abs, followed by a mixture of goat anti-rat and goat anti-mouse IgG-coupled magnetic beads. A total of 2 10 4 purified cells were mixed with 5 10 5 3000 rad irradiated compatible (SJL/J) splenocyte feeders, stimulated, and analyzed as above. Isolation of naive CD4 T cells was performed as described (39). Briefly, splenocytes and lymph node cells were mixed, and red cells lysed with hypotonic buffer. Cells were then stained with a mixture of CD8 and NK-specific Abs and negatively selected using a mixture of goat anti-mouse and goat antirat IgG-coupled beads, as above. These cells were then labeled with FITCconjugated CD45RB and CyChrome-conjugated CD44 and flow cytometrically sorted to collect naive (CD44 low, CD45RB high ) cells. Alternatively, lymph node cells were stained with Abs specific for CD4, CD44, and CD45RB, and gated CD4 T cells were sorted into naive and memory (CD44 high, CD45RB low ) populations. Th cell differentiation and cytokine ELISA Naive T cells were isolated as above and assayed for Th1 and Th2 differentiation as described (39). Briefly, SJL/J splenocyte feeders were depleted of T and NK cells by incubation with Y-19 and HB191 Abs and rabbit complement and 3000 rad irradiated. A total of 5 7.5 10 5 feeders was mixed with an equal number of naive CD4 T cells and 30 U/ml human ril-2, and incubated in 2C11- or IA.B2-coated 48-well plate wells. To promote differentiation into Th1 cells, 11B11 (anti-il-4) Ab was added with 3.5 ng/ml murine ril-12. To promote Th2 differentiation, XMG1.2 (anti-ifn- ) Ab and 4000 U/ml murine ril-4 were added. Cells were cultured for 4 days. To remove XMG1.2 Ab to permit analysis of IFN- production, cells from each well were washed with medium four times and replated with 30 U/ml human ril-2 in 48-well plate wells coated with the same stimulating Ab as initially present. Approximately 20 h later, supernatants were harvested and analyzed for IL-5 and IFN- production by a sandwich-type ELISA using manufacturer s protocol (PharMingen). Results Transgenic mice with a chimeric class II MHC- receptor A chimeric construct was produced by linking the class II MHC IA s or IA s extracellular and transmembrane domains with the cytoplasmic domain of the murine TCR -chain. In the case of the IA s chain, covalent peptide was linked to the N terminus as described in Materials and Methods (26, 27) (Fig. 1). Transfection of these constructs under the control of the Moloney murine leukemia virus long terminal repeat into the 002 T cell hybridoma line verified appropriate association of the - and -chain chimeric constructs. Transfectants were class II-positive by surface staining and flow cytometry and produced IL-2 upon chimeric receptor crosslinking (data not shown). These constructs were then subcloned into the phcd2-va vector, placing them under the control of the FIGURE 1. Amino acid sequence of IA s - (I) and IA s - (II) chimeric constructs. Regions correspond to: A, IA k leader sequence (aa 1 31); B, IA d 1 domain (aa 28 31); C, myelin basic protein peptide (89 101) (aa 34 46); D, peptide linker (aa 47 61); E, IA s 1, 2, transmembrane domains (aa 62 282); F, -chain cytoplasmic domain (aa 285 397); G, IA s leader sequence, 1, 2, transmembrane domains (aa 1 223); H, -chain cytoplasmic domain (aa 226 338). Unlisted sequence is as published (48 50), except complete leader sequence for IA s is MPCSRALILGVLALTTMLSLCGG. human CD2 promoter and locus control region (LCR), and coinjected into (B6xC3H)F 2 day 1 embryos to generate transgenic mice. The phcd2 and LCR have been shown to produce integration site-independent lymphoid-specific expression in transgenic mice (29, 30). Because murine T cells do not express endogenous class II molecules, expression of the transgenic construct could be monitored with the murine class II MHC-specific Ab IA.B2. The 25S89P transgenic line constitutively expresses class II on both CD4 and CD8 T cells. A similar flow cytometric staining profile is observed among peripheral blood, lymph node, and splenic T cells (Fig. 2, A and B, and data not shown). A broad distribution of expression levels is seen among T cells. The level of transgene expression could not be associated with memory/naive or activation phenotype, as measured with the CD44, CD45RB, CD62L, CD69, and CD25 Abs. T cell development does not appear affected by the IA s - transgene. Flow cytometric analysis of transgenic or nontransgenic thymocytes stained with CD4 and CD8 does not show any developmental perturbation. Additionally, similar numbers and classes (CD4, CD8, naive/memory) of peripheral T lymphocytes are observed in positive and negative mice (data not shown). Further, FIGURE 2. Flow cytometric and Western blot analyses of chimeric receptor. Histogram analysis of chimeric receptor expression on 25S89P transgenic lymph node cells stained with IA.B2 (anti-class II MHC)-FITC and either CD4-PE or CD8-PE. IA.B2 expression on CD4 gated cells (A) or CD8 gated cells (B) are shown. C, Equivalent numbers of purified T cells from transgenic or nontransgenic mice were lysed, and equivalent amounts of whole cell lysate separated on an 8 12% denaturing SDS- PAGE gradient gel. The Western blot was probed with polyclonal anti- TCR- Ab, demonstrating chimeric receptor as well as native. Positions of m.w. markers are shown. Assignment of - and -chains of the chimeric receptor are based on calculated m.w. The 28-kDa band is unidentified cross-reactive material, also observed using preimmune serum.
5934 -CHAIN-MEDIATED PRIMARY T CELL ACTIVATION FIGURE 3. IA s - cross-linking can induce calcium mobilization in primary T cells. T cells isolated from 25S89P transgenic mice were loaded with the calcium sensitive fluorochrome, Fluo-3, stimulated by Ab cross-linking, and changes in intracellular calcium were monitored in individual cells using a video laser cytometer. The cells were preincubated with 5 g/ml of the indicated primary Abs, and, at the first arrow, goat anti-rat (GAR 100 g/ml) Ab was added. Each graph indicates the pattern of calcium mobilization in a field of 45 50 cells, where each line represents the average fluorescent intensity of an individual T cell over time. The second arrow indicates the addition of ionomycin (666 ng/ml). Control experiments indicated that cross-linking with the primary Abs alone was not sufficient to induce significant calcium mobilization. TCR and CD4 and CD8 coreceptor expression in the thymus and periphery are comparable to that of nontransgenic littermate controls, demonstrating that the transgenic receptor was not influencing levels of these signaling molecules. A slightly increased level of chimeric receptor is observed on CD8 T cells when compared with CD4 T cells. By comparing staining intensity of CD4 and CD8 T cells, it can be estimated that mean expression of IA s - receptors on CD4 T cells is 75% of that on CD8 T cells, and median expression is 47% of that on CD8 T cells. Although the cause of the differences in expression cannot be ascertained with certainty, because the hcd2 promoter/ LCR has been demonstrated to be integration site-independent (29, 30) and expresses well on all mature cells, we suspect that pairing of the and chimeric receptor chains may have different efficiencies in different cells. Alternatively, the turnover rate of the chimeric receptor may differ in different cells. Western blot analysis of whole cell lysate from transgenic T cells demonstrates that the total amount of class II chimeric and endogenous -chain molecules on transgenic T cells are similar (Fig. 2C). Thus, although single cells express heterogeneous levels of chimeric receptor, the average total cellular level approximates that of native -chain. Chimeric IA s - is able to activate primary T lymphocytes To determine whether the -chain cytoplasmic domain is sufficient to transduce signals into primary T lymphocytes, calcium flux in response to receptor cross-linking was measured using the class II MHC-specific Ab IA.B2 or the control CD3 -specific Ab 2C11. Fig. 3, A C shows single cell intracytoplasmic calcium tracings of purified primary T lymphocytes after either IA.B2 or 2C11 crosslinking. Both IA.B2 and 2C11 can induce a significant rise in intracellular calcium levels. As compared with 2C11 cross-linking, a smaller fraction of cells flux calcium with IA.B2 cross-linking. Approximately 44% of T cells fluxed calcium in response to IA.B2, compared with 92% when activated with 2C11. Further, the IA.B2-stimulated cells are less synchronized in the time delay to fluxing. These differences likely reflect the variable cell surface expression of the chimeric receptor, with 50% of the cells in the mice studied containing insufficient receptor level to initiate a calcium flux. Fig. 3, D F shows that co-cross-linking CD4 with the chimeric receptor enhances the level of calcium fluxed and diminishes the time to flux, but does not alter the percentage of T cells that flux calcium (46% for IA.B2 and 88% for 2C11). The enhancement of chimeric receptor signal by simultaneous CD4 crosslinking is similar to the ability of CD4 cross-linking to enhance 2C11-induced calcium mobilization. This suggests that CD4 can function in cooperation with the class II receptor s ITAMs much as it does with the TCR/CD3 complex, possibly by the recruitment of lck to the chimeric receptor (40). Analysis of proliferative response to either TCR or IA s - crosslinking demonstrates that the signal generated by the chimeric receptor is indeed functional (Fig. 4). Whole splenocytes, as well as enriched CD4 or CD8 T cells, proliferate vigorously after chimeric receptor cross-linking. This implies that the -chain cytoplasmic domains are adequate to promote full activation of primary T lymphocytes. There was no a priori reason to suspect that IA s - -mediated signaling may be dependent on other TCR elements. Nevertheless,
The Journal of Immunology 5935 FIGURE 4. Proliferative response of splenocytes from 25S89P transgenic (filled, open circles) and nontransgenic mice (filled, open squares). Response of whole splenocytes to plate-bound 2C11 (A) or IA.B2 (B) stimulation is shown. B-depleted splenocytes showed a similar response (data not shown). Isolated CD4 or CD8 T cells (C) were purified from lymph node cells using magnetic bead negative selection to 90% purity and proliferative response assessed as described in Materials and Methods. because both native TCR as well as the chimeric receptor are expressed on the transgenic T cells, it was necessary to exclude this possibility before concluding that the IA s - receptor was truly responsible for T cell activation. To do this, we analyzed tyrosine phosphorylation of the native -chain in response to cross-linking of the chimeric receptor. If signaling through the chimeric receptor requires an association with native TCR, then cross-linking of the chimeric receptor with the IA.B2 Ab should induce hyperphosphorylation of the native TCR -chain. Purified T cells were thus activated with either 2C11 or IA.B2 Abs in the presence or absence of co-cross-linking with anti-cd4. -chain was immunoprecipitated and probed with either anti-phosphotyrosine or -specific Abs. Fig. 5A shows that, whereas activation through CD3 or CD3/ CD4 readily induced native hyperphosphorylation, a baseline phosphorylation state was seen after activation through IA s -, with or without CD4. This implies that the chimeric receptor signals independently of TCR. If the chimeric receptor signaled through the TCR, then phosphorylation of native should have been apparent. In a parallel fashion, activation through the IA s - receptor but not the TCR induces phosphorylation of IA s -. At least two appropriately sized phosphorylated bands are induced upon crosslinking transgenic T cells with either class II-specific Ab or a mixture of class II and CD4/CD8-specific Abs (Fig. 5B). These bands, consistent with phosphorylated class II-, are not induced following CD3 cross-linking. Induction of these bands is not seen when nontransgenic T cells are cross-linked with class II or CD3-specific Ab (data not shown). The increased receptor phosphorylation observed when CD4/CD8 coreceptor is cross-linked together with IA s - parallels the augmented Ca flux seen with coreceptor cross-linking (Fig. 3). This demonstrates an enhancement of signaling in the presence of coreceptor cross-linking. Additional phosphorylated species comigrating with Ig heavy chain (IgH) may also be present, but interference from IgH bands prevent positive identification. Thus, this data shows that the class II- chimera is appropriately phosphorylated upon cross-linking and further confirms the functional independence of the chimeric receptor and TCR. To determine whether the Ab-mediated stimulation used in these phosphorylation studies was sufficient to initiate downstream signaling, we analyzed stimulation-induced ZAP-70 recruitment to -chain (chimeric or native) (Fig. 5C). Stimulated samples were immunoprecipitated with -specific Ab capable of binding both native and chimeric class II- molecules and probed with a phosphotyrosine-specific Ab. Recruitment was detectable in response to anti-cd3 or anti-class II Ab-mediated stimulation in the presence of anti-cd4/cd8 Abs. This demonstrates that the chimeric receptor can initiate downstream signaling events. FIGURE 5. Fidelity of tyrosine phosphorylation of TCR-associated and class II- chimera following Ab cross-linking of CD3 or class II, respectively. A, Cross-linking of IA s - chimera does not result in phosphorylation of the endogenous -chain. Purified 25S89P transgenic and nontransgenic T cells were stimulated by Ab cross-linking for 2 min, as described in Fig. 3 and Materials and Methods. 1% Nonidet P-40 soluble cell lysates were immunoprecipitated with antiserum and assessed for tyrosine phosphorylation by Western blot after 12% SDS-PAGE. The arrows indicate mobility of the fully saturated (p21) and partially saturated (p18) phosphozeta isoforms. The same blot was then stripped and reprobed for total protein. B, Purified 25S89P transgenic T cells were stimulated as in A and lysates immunoprecipitated with anti-class II Ab. Tyrosine-phosphorylated class II- was assessed by Western blot after 8 12% gradient SDS-PAGE. The same blot was then stripped and reprobed for as in A. The identity of tyrosine-phosphorylated and total class II- was made by comparison with nontransgenic T cells stimulated in an identical fashion (not shown). C, Cells were stimulated by Ab cross-linking and immunoprecipitated with anti- as in A. The presence of an associated tyrosine-phosphorylated p70 protein (presumed ZAP-70) was detected with Western blot.
5936 -CHAIN-MEDIATED PRIMARY T CELL ACTIVATION FIGURE 6. Receptor cocapping analysis of IA s - and TCRs. The study was performed as described in Materials and Methods. Cells were capped with IA.B2, followed by goat anti-igg-fitc, fixed, and stained with follow-up 2C11-biotin and avidin-texas Red (A C). Alternatively capping with 2C11 was followed by fixation and staining for class II MHC (D F). Photomicrographs show representative cells with differential interference contrast (A and D), fluorescent (FITC) filtered for capping Ab (B and E), or fluorescent (Texas-Red) filtered for follow-up Ab (C and F). The small amount of clumpiness observed with the follow-up Ab (C and F) was not apparent when fluorochrome was directly conjugated to the follow-up Ab (data not shown). Although the phosphorylation studies all suggest that IA s - acts independently of the TCR, we further verified this by directly analyzing association of the receptors. Two types of studies were performed. First, we analyzed TCR and IA s - coimmunoprecipitation. If -chain associates with the IA s - receptor, immunoprecipitation of IA s - may be expected to coimmunoprecipitate. No such coimmunoprecipitation was seen, implying that the two molecules are not associated on the cell surface (data not shown). Second, we analyzed TCR and IA s - cocapping. Cross-linking of the TCR in metabolically active cells has been shown to result in receptor capping that is visible microscopically with fluorescently labeled Abs (37). In metabolically inactive cells, such as those kept cold or fixed, no capping is seen. If the IA s - receptor signals via the TCR, capping of the IA s - receptor should cocap the TCR. When transgenic T cells were incubated with either 2C11 or IA.B2 and then cross-linked with a secondary Ab, capping of the T cell or IA s - receptor was indeed seen (data not shown). When such capped cells were then fixed and incubated with Ab to the complementary receptor (IA.B2 or 2C11, respectively), it was apparent that the receptors capped independently (Fig. 6). These studies confirm that the IA s - receptor is acting independently of the TCR. Together with the biochemical and proliferation studies shown in Figs. 3 5, it can be concluded that the -chain cytoplasmic domain, FIGURE 7. Proliferative response of naive and memory T cells to IA s - stimulation. CD4 T lymphocytes were separated into CD45RB high, CD44 low (naive) and CD45RB low, CD44 high (memory) populations by flow cytometric sorting. A total of 2 10 4 sorted cells were stimulated in the presence of 5 10 5 3000 rad irradiated SJL/J splenocytes using 5 g/ml plate-bound IA.B2, 5 g/ml plate bound 2C11, or without coating. Means of duplicate samples from a representative experiment are shown. exclusive of the remainder of the TCR, is able to functionally stimulate primary T lymphocytes. Both naive and memory phenotype T cells are equivalently activated through the IA s - receptor Some studies have shown that memory/preactivated T lymphocytes can be activated with substantially less intense stimuli than naive T lymphocytes (41). To determine whether the IA s - construct was limited in its ability to signal, only stimulating memory T cells, CD4 memory (CD44 high, CD45RB low ) and naive (CD44 low, CD45RB high ) T cells were flow cytometrically purified. These purified populations were then stimulated with either IA.B2 or 2C11 in the presence of APC feeders, and proliferative response was measured. Fig. 7 shows that naive and memory cells are equivalently stimulated. The cytoplasmic domain is thus capable of even activating naive T lymphocytes. The -chain cytoplasmic domain is sufficient to promote T cell maturation into CTL, Th1, and Th2 T lymphocytes We next analyzed whether stimulation through the IA s - receptor could promote the differentiation of mature primary T lymphocytes FIGURE 8. Differentiation of 25S89P T cells into CTL. Splenocytes from 25S89P mice were stimulated with 5 g/ml plate-bound 2C11 (open symbols) or IA.B2 (filled symbols) Ab for 5 days, harvested, and assayed in a directed cytolysis assay against 2C11-coated CHO-B1 (Fc RIIB1- transfected CHO) cells (triangles), control uncoated CHO-B1 cells (squares), or control 2C11-coated untransfected CHO cells (circles) as described in Materials and Methods.
The Journal of Immunology 5937 FIGURE 9. Differentiation of naive 25S89P transgenic T cells into Th1 and Th2 subsets. Naive T cells were purified as described in Materials and Methods, mixed with T and NK cell-depleted irradiated splenocyte feeders, and stimulated for 4 days with plate-bound IA.B2 (A and C) or 2C11 (B and D) in the presence of IL-12 and anti-il-4 (Th1 conditions) or IL-4 and anti-ifn- (Th2 conditions). They were then washed and stimulated again with the same plate-bound Ab for 20 24 h. Supernatants were harvested and assayed for IL-5 and IFN- as representative Th2 and Th1 cytokines. into different effector populations. To analyze differentiation into CTL, T cells from transgenic or nontransgenic mice were preactivated for 5 days with either IA.B2 or 2C11. This time frame has been shown sufficient to allow differentiation into functional CTL (42). At the end of this period, bulk cytolytic ability was measured in a directed lysis assay, using Fc RIIB1-transfected 2C11-coated CHO cells (CHO-B1 cells) as targets. Fig. 8 shows that either 2C11 or IA.B2 is capable of promoting transgenic T lymphocyte maturation into CTL, with no apparent difference when the T cells were matured by stimulation through the IA s - chimeric receptor vs the TCR. As expected, nontransgenic T cells develop into CTL only after activation with 2C11. Stimulation through IA s - is thus sufficient to promote primary T cell differentiation into CTL. To determine whether IA.B2-induced activation could likewise mediate differentiation into Th1 or Th2 T cells, CD4 T cells from transgenic or nontransgenic mice were stained with CD45RB and CD44 and flow cytometrically sorted to collect naive (CD45RB high, CD44 low ) T lymphocytes. These cells were then stimulated with plate-bound IA.B2 or 2C11 Abs in the presence of irradiated T and NK cell-depleted APCs for 4 days. To promote Th1 or Th2 development, IL-12 with anti-il-4 or IL-4 with anti-ifn- was added respectively to different cultures. The cells were then washed and stimulated for an additional day using either IA.B2 or 2C11 as a stimulus. IFN- and IL-5 production were measured as indicators of Th1 and Th2 development, respectively. Fig. 9 shows that both transgenic and nontransgenic cells matured into Th1 and Th2 effectors after stimulation with 2C11. Yet, only transgenic cells developed into Th1 and Th2 cells after stimulation with IA.B2. Thus, in the appropriate cytokine environment, the IA s - chimeric receptor is able to induce differentiation into either Th1 or Th2 effector cells. Discussion The role of specific ITAMs in T cell signaling and differentiation is poorly understood. A knowledge of how ITAMs transduce signals has assumed increasing clinical relevance as chimeric molecules containing TCR signaling domains are being developed for clinical use (22, 23, 43). We have analyzed how a subset of TCR ITAMs, those contained within the cytoplasmic domain of the TCR -chain, can mediate T cell maturation and function. We show that cytoplasmic, dimerized using the MHC class II IA s extracellular and transmembrane domains, can induce activation and differentiation of primary CD4 and CD8 T lymphocytes. Transgenic T cells flux calcium, proliferate, and mature into CTL, Th1, and Th2 subsets normally. Further, we show that the IA s - chimeric receptor does not associate with and acts independently of native TCR molecules. If the ITAMs are adequate for maturation of primary T cells, then the remaining TCR ITAMs, present on the,, and CD3 chains, appear redundant. Indeed, in T cell hybridomas, cross-linking single ITAMs is sufficient to induce IL-2 production (15, 16). Signal strength is increased with increased numbers of ITAMs. Other studies demonstrate that, in the absence of any ITAMs, T cells mature in the thymus and are grossly normal, using proliferation in response to anti-cd3- or Con A stimulation as a measure (13). Therefore, either the -chain, as shown in this study, or, alternatively, all other ITAMs seem adequate for T cell development and/or activation. If only one or a few ITAMs are sufficient to signal T cells, why are so many present in the TCR? Two possibilities exist. The many ITAMs may act to amplify the TCR signal. Alternatively, they may act to mediate independent signals by coupling with different signal transduction molecules. Our results are more consistent with the former possibility. The ITAMs, exclusive of the other components of the TCR, induce functional differentiation of mature T cells into CTL, Th1, and Th2 cells. This concept is complemented by other studies showing normal thymic development and thymocyte signaling in T cells lacking all or some TCR ITAMs (13, 14, 25). Therefore, both thymic and mature T cell differentiation does not require an intact complement of TCR ITAMs. One other study has extensively analyzed peripheral T cell responses in transgenic mice expressing chimeric proteins with the signaling domain (24). The results, using a human CD3 -specific Fv- chimera, are in direct contrast to those of this study. Primary T cells failed to flux calcium or proliferate in response to crosslinking of the Fv- chimeric receptor. Memory T cells or T cells preactivated with anti-cd3 did respond. This study was interpreted to show that the -chain ITAMs were unable to activate primary resting T cells. Implicit in this interpretation is that the -chain ITAMs lack structural motifs required for linking to critical downstream signaling molecules. This interpretation is not certain, however. The Fv- construct contained the complete -chain transmembrane sequence (24, 44). Because this sequence is sufficient for -chain dimerization (7), the construct would be expected to heterodimerize with native -chain expressed on the T cells of these transgenic mice. As a result, the Fv- chimera could potentially associate with native TCR on T cells. Because cross-linking of the chimeric receptor would also cross-link TCR, the absence of primary T cell stimulation cannot be attributed exclusively to ineffective signaling by -chain ITAMs. In contrast, in the chimeric construct used in our study here, the class II transmembrane domain was used. This would not be expected to, and based on tyrosine phosphorylation and receptor cocapping studies shown here, does not associate with the TCR. Why cross-linking failed to stimulate primary T cells in the Fv- transgenic mice as it does here is not clear. Indeed, it may be anticipated that linking the chimeric receptor with the TCR should enhance T cell signaling. One possibility is that the Fv- -TCR
5938 -CHAIN-MEDIATED PRIMARY T CELL ACTIVATION complex transduced an antagonistic rather than agonistic signal. Alternatively, the signal transmitted by the Fv- was too weak in primary T cells to generate a functional response, but was effective in preactivated cells that have a decreased activation threshold. Resolving why this chimeric -chain molecule failed to signal primary T cells may be important in the future design of chimeric constructs for clinical use. Further studies will be required to address this issue. Although the idea that multiple ITAMs serve primarily to amplify signal is consistent with this study, it still must be reconciled with data suggesting that different ITAMs transmit functionally distinct signals. Data supporting this latter idea come primarily from two sources. First, in vitro assays studying peptide and protein association show preferential SH2 domain binding with certain ITAMs (9, 45). However, these studies may not adequately reflect the complex internal environment during cellular signaling. The interactions of signal-transducing proteins with ITAMs within the cell may, in fact, differ from the conditions used in these studies. Other data implying a distinct role for different ITAMs comes from studies of ITAM-containing constructs expressed in T cell hybridomas. Yet again, hybridomas may not adequately represent signaling in primary T cells. Indeed, studies with chimeric constructs containing and ITAMs expressed in hybridomas showed differential protein phosphorylation (46). Yet, these constructs did not display such differences when expressed as transgenes in mouse T cells (25). Despite these caveats, it must be noted that our results do not necessarily contradict the idea that individual ITAMs preferentially associate with select signal transduction molecules. However, each ITAM seems not to have a unique and critical role in T cell signaling. A similar signal transduction complex may form regardless of the set of ITAMs present. The efficiency of its formation, however, may depend on both the subset and number of ITAMs available. Thus, whereas it seems feasible that there is some substrate specificity among different ITAMs, such specificity may be of limited importance in primary T cell signaling. Indeed, in this regard it is noteworthy that CTL lines can be activated by cross-linking of chimeric molecules containing syk (47). Therefore, in select circumstances, cross-linking of single molecules downstream of ITAM phosphorylation may be able to induce complete T cell activation. Further studies will be necessary to clarify the composition and mechanism of generation of the TCR signaling complex in response to limited sets of ITAMs. Many studies have documented the role of signaling molecules besides the TCR in T cell activation and differentiation. The in vitro studies conducted here provided free access of costimulatory molecules presented by APCs to the T cells. How receptors with limited sets of ITAMs will function in situations where costimulation is limiting remains to be determined. Likewise, although the functional studies shown here demonstrate a roughly equivalent capacity of the TCR and IA s - receptor to stimulate T cells, the biochemical and calcium flux data also demonstrate a clearly enhanced signal when CD4 and/or CD8 coreceptor is also crosslinked. Whereas TCR stimulation will, by its nature, accompany coreceptor stimulation, this is not necessarily true of chimeric constructs. The importance of coreceptor signaling is well established and needs to be considered in the design of chimeric constructs for therapeutic purposes. Thus, although these studies support the use of the -chain cytoplasmic domain for the creation of fully functional chimeric receptors on primary T lymphocytes, newer construct designs should consider methods that may enhance signaling by coengaging relevant coreceptor molecules. Acknowledgments We thank Debbie Butkus, Cindy Hughes, and Lisa Chuba for their assistance in generating and maintaining transgenic mice; Tom Taylor for assistance with cell sorting; and Fran Manzo for assistance in manuscript preparation. References 1. Clevers, H., B. Alarcon, T. Wileman, and C. Terhorst. 1988. The T cell receptor/ CD3 complex: a dynamic protein ensemble. Annu. Rev. Immunol. 6:629. 2. Ashwell, J., and R. Klausner. 1990. Genetic and mutational analysis of the T-cell antigen receptor. Annu. Rev. Immunol. 8:139. 3. Weiss, A., and D. Litman. 1994. Signal transduction by lymphocyte antigen receptors. Cell 76:263. 4. Weiss, A. 1993. T cell antigen receptor signal transduction: a tale of tails. Cell 73:209. 5. Cambier, J. 1995. New nomenclature for the RETH motif (ARH1/TAM/ARAM/ YXXL). Immunol. Today 16:110. 6. Isakov, N. 1997. Immunoreceptor tyrosine based activation motif (ITAM) a unique module linking antigen and Fc receptors to their signaling cascades. J. Leukocyte Biol. 61:6. 7. Rutledge, T., P. Cosson, N. Manolios, J. Bonifacino, and R. Klausner. 1992. Transmembrane helical interactions: chain dimerization and functional association with the T cell receptor. EMBO J. 11:3245. 8. Sloan-Lancaster, J., and P. Allen. 1996. Altered peptide ligand induced partial T cell activation: molecular mechanisms and role in T cell biology. Annu. Rev. Immunol. 14:1. 9. Kimura, T., H. Kihara, S. Bhattacharyya, H. Sakamoto, E. Appella, and R. Siraganian. 1996. Downstream signaling molecules bind to different phosphorylated immunoreceptor tyrosine-based activation motif (ITAM) peptides of the high affinity IgE receptor. J. Biol. Chem. 271:27962. 10. Rozdzial, M., B. Malissen, and T. Finkel. 1995. Tyrosine-phosphorylated T cell receptor chain associates wih the actin cytoskeleton upon activation of mature T lymphocytes. Immunity 3:623. 11. Jensen, W., C. Pleiman, P. Beaufils, A. Wegener, B. Malissen, and J. Cambier. 1997. Qualitatively distinct signals through T cell antigen receptor subunits. Eur. J. Immunol. 27:707. 12. Shores, E., K. Huang, T. Tran, E. Lee, A. Grinberg, and P. Love. 1994. Role of TCR chain in T cell development and selection. Science 266:1047. 13. Shores, E., T. Tran, A. Grinberg, C. Sommers, H. Shen, and P. Love. 1997. Role of multiple T cell receptor (TCR)- chain signaling motifs in selection of the T cell repertoire. J. Exp. Med. 185:893. 14. van Oers, N., P. Love, E. Shores, and A. Weiss. 1998. Regulation of TCR signal transduction in murine thymocytes by multiple TCR -chain signaling motifs. J. Immunol. 160:163. 15. Romeo, C., M. Amiot, and B. Seed. 1992. Sequence requirements for induction of cytolysis by the T cell antigen/fc receptor chain. Cell 68:889. 16. Irving, B., A. Chan, and A. Weiss. 1993. Functional characterization of a signal transduction motif present in the T cell receptor chain. J. Exp. Med. 177:1093. 17. Romeo, C., and B. Seed. 1991. Cellular immunity to HIV activated by CD4 fused to T cell receptor polypeptides. Cell 64:1037. 18. Irving, B., and A. Weiss. 1991. The cytoplasmic domain of the T cell receptor chain is sufficient to couple to receptor associated signal transduction pathways. Cell 64:891. 19. Wels, W., D. Moritz, M. Schmidt, M. Jeschke, N. Hynes, and B. Groner. 1995. Biotechnological and gene therapeutic strategies in cancer treatment. Gene 159: 73. 20. Tran, A., D. Zhang, R. Byrn, and M. Roberts. 1995. Chimeric receptors direct human natural killer effector function to permit killing of NK resistant tumor cells and HIV infected T lymphocytes. J. Immunol. 155:1000. 21. Letourneur, F., and R. Klausner. 1991. T cell and basophil activation through the cytoplasmic tail of T cell receptor family proteins. Proc. Natl. Acad. Sci. USA 88:8905. 22. Yang, O., A. Tran, S. Kalams, R. Johnson, M. Roberts, and B. Walker. 1997. Lysis of HIV-1 infected cells and inhibiton of viral replication by universal receptor T cells. Proc. Natl. Acad. Sci. USA 94:11478. 23. Abken, H., A. Hombach, U. Reinhold, and S. Ferrone. 1998. Can combined T-cell and antibody-based immunotherapy outsmart tumor cells? Immunol. Today 19:2. 24. Brocker, T., and K. Karjalainen. 1995. Signals through T cell receptor chain alone are insufficient to prime resting T lymphocytes. J. Exp. Med. 181:1653. 25. Shinkai, Y., A. Ma, H. Cheng, and F. Alt. 1995. CD3 and CD3 cytoplasmic domains can independently generate signals for T cell development and function. Immunity 2:401. 26. Kozono, K., J. White, J. Clements, P. Marrack, and J. Kappler. 1994. Production of soluble class II proteins with covalently bound single peptides. Nature 369: 151. 27. Ignatowicz, L., G. Winslow, J. Bill, J. Kappler, and P. Marrack. 1995. Cell surface expression of class II MHC proteins bound by a single peptide. J. Immunol. 154:3852. 28. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1989. In Current Protocols in Molecular Biology. John Wiley and Sons, New York, pp. 3.7.1. 3.7.3.