of t cell clone to a large panel of altered peptide

Response of a Human T Cell Clone to a Large Panel of Altered Peptide Ligands Carrying Single Residue Substitutions in an Antigenic Peptide

Characterization and Frequencies of TCR Agonism and TCR Antagonism With or Without Partial Activation’

Yu-Zhen Chen, Sho Matsushita, and Yasuharu Nishimura*

A CD4+ human T cell clone YN5-32 recognized a streptococcal M12p54-68 peptide in the context of HLA-DR4 and produced a large amount of IFN-y. We investigated responses of YN5-32 to 156 independent analogue peptides carrying single residue substitutions at residues 57 (position 1 (PI)) to 65 (p9) of the peptide. Approximately 30% of analogues at either Leu5’ (pl), Ala6’ (*), or As@ (p6) residues exhibited TCR agonism to stimulate various magnitudes of proliferative responses in the T cell clone, and analogues exhibiting TCR antagonism are rare in these three residues. In analogues at either C I U ~ ~ (p2), Gln5’ (p3), Tyr6’ (p5), or G I u ~ ~ (p7) residue, 30 to 50% exhibited TCR antagonism. About 10% of analogues at G I u ~ ~ (p2) or Tyr6’ (p5) stimulated proliferative responses, while 30 to 50% of analogues at Cln59 (p3) or G I d 3 (p7) did so. Some of these TCR antagonistic analogues carrying relatively conservative amino acid substitutions partially activated the T cells to induce large increases in size and expression levels of CD4, CD1 l a (LFA-l), CD28, CD49d (VLA-4), and CD95 (Fas), and small increases in CD25 and CD44 expressions on the cell surface. None of the partially activating antagonistic analogues induced IFN-y production or anergy in T cells. Analogues with replacements of acidic amino acids at either Led4 (p8) or Ser6’ (p9) residue had dominant negative effects on T cell proliferation. Thus, altered peptide ligands with single residue substitutions in the antigenic peptide frequently stimulated the human T cell clone, in at least three different ways to exhibit agonism, antagonism, and antagonism with partial activation. Frequencies of analogue peptides exhibiting these three different effects on the T cell clone differed depending on the residue of the peptide substituted. Altered T cell responses induced by analogue peptides of a T cell epitope provide a system to analyze activation signals mediated by TCR, and to manipulate T cell responses. The lournal of Immunology, 1996, 157: 3783-3790.

T he HLA-DR molecule is a heterodimeric membrane protein consisting of monomorphic a- and highly polymor- phic @chains, and is expressed on B cells, APC, and ac-

tivated T cells. The HLA-DR molecule has a peptide-binding groove on top of the molecule and binds antigenic peptides pro- cessed by APC to present them to CD4+ T cells (see Ref. 1 for review). Positions and characteristics of amino acid residues important for HLA-DR binding and for recognition by TCR were determined in many combinations of HLA-DR molecules and peptides through direct binding assay between purified DR molecules and analogue peptides carrying single residue substitutions in DR- binding peptide (2-9), sophisticated phage random peptide library (10, 1 l), or functional assay for capacity of peptides to stimulate T

Divislon of Immunogenetics, Department of Neuroscience and Immunology, Kumamoto University Graduate School of Medical Sciences, Kumamoto, Japan

Received for publication August 7 , 1995. Accepted for publication August 13, 1996.

The costs of publication of this article were defrayed in part by the payment of

accordance with 18 U.S.C. Section 1734 solely to indicate this fact. page charges. This article must therefore be hereby marked advertisement in

05272104 from the Ministry of Education, Science, Sports, and Culture of Japan; ’ This work was supported in part by Grants-in-Aid 03452276, 052781 18, and

a Research Grant for Intractable Diseases from the Ministry of Health and Wel- fare, lapan; lchiro Kanehara Foundation: Terumo Life Science Foundation: Japan Rheumatism Foundation; and Mochida Memorial Foundation.

Address correspondence and reprint requests to Dr. Yasuharu Nishimura, Di- vision of Immunogenetics, Department of Neuroscience and immunology, Kumamoto University Graduate School of Medical Sciences, Honjo 2-2-1, Kumamoto 860, Japan.

Copyright 0 1996 by The American Association of Immunologists

cell responses (12-1s). Recent crystallographic analyses of the DRI molecule complexed with either self peptides (16) or an influenza hemagglutinin peptide (17) elucidated the physical basis of interaction between DR molecule and peptides. According to these results, at least five DR anchor residues on the DRI-binding peptide and five corresponding independent pockets of DR molecule that accommodate side chains of DR anchor residues of the peptide were identified. Sixty-five percent of the surface of peptide made contact with the DR molecule, and the remaining peptide was ac- cessible to solvents, being recognized by TCR.

CD4+ T cells usually recognize non-self peptides in the context of self HLA-DR molecules, and recognition and responses of T cells seemed to be an odo f f phenomena. However, recent findings in mice, examined by utilizing analogue peptides carrying single residue substitutions in antigenic peptides, revealed that T cell clones recognized these altered peptide ligands. Altered peptide ligands carrying amino acid substitutions at the TCR contact residues on the peptides also induced T cell nonresponsiveness through TCR antagonism (18-20) or anergy induction as a consequence of partial or incomplete activation (21, 22), and some- times induced dissociation between proliferative response and cytokine production (23, 24). These altered T cell responses to various altered peptide ligands were evidenced in both mice (19, 21-24) and humans (18, 20, 25, 26). Some TCR antagonistic analogue peptides induced partial activation of T cells to increase cell size and expression levels of CDl la (LFA-I) and CD25 (IL-2R) on surface of the T cell, and induced anergy in the T cell clone (21). This partial activation of T cells by TCR antagonistic altered

0022-1 767/96/$02.00

3784 HUMAN T CELL RESPONSE T O ALTERED PEPTIDE LIGANDS

peptide ligands was noted in murine T cells, and such events in human T cells have remained to be determined.

We analyzed responses of a human T cell clone to a large panel of analogue peptides derived from an antigenic peptide presented by DR4 (DRB1*0406), the objective being to investigate frequencies and characteristics of TCR agonism and TCR antagonism, with or without partial activation. We chose the DRB1*0406 molecule as an Ag-presenting molecule because it is strongly associated with susceptibility to insulin autoimmune syndrome, and because we have already identified motif of the peptides bound to the DRB 1 *0406 complex (7). Many analogue peptides carrying single residue substitutions in the antigenic peptide exhibited full agonism for the T cell clone to induce various magnitudes of proliferative response or TCR antagonism to inhibit proliferative response of the T cell clone to the wild-type peptide. Some TCR antagonistic analogue peptides partially stimulated T cells to induce increases in cell size and expression levels of several cell surface molecules without inducing cell proliferation, IFN-y production, or anergy in the T cell clone. Marked differences in frequencies of analogue peptides exhibiting three different effects on the T cell clone were evident among residues of peptide substituted.

Materials and Methods Synthesis of peptides

Thirty of 19 to 25 amino-acid-long overlapping peptides corresponding to the primary sequence of MI2 protein (27), an M protein derived from a group A p hemolytic streptococcal strain 12, were synthesized in our lab- oratory. Among these peptides, the M12p50-72 ('TKMLNRDLE QAYNELSGEAHKDA") corresponds to amino acid residues 50 to 72 of the MI2 protein. All of the peptides used, including the truncated form of M12p50-72 peptide and its analogues carrying only one residue substitution, used were synthesized using a solid-phase simultaneous multiple peptide synthesizer PSSM-8 (Shimadzu, Kyoto, Japan), based on Fmoc strat- egy. All peptides were purified by reverse-phase HPLC (Millipore, Waters, Milford, MA), and accuracy of amino acid sequences was confirmed using a protein sequencer (PPSQ-10; Shimadzu). Binding capacities to the purified DRB I *0406 molecules of several peptides were determined by competitive binding inhibition assay, as described elsewhere (7), with minor modifications in which 20 nM of radioiodinated 0405BP3 (GSTVFDN LPNPEIDGDYYGW), a good binder not only to DRBl*0405 hut also to closely related DRB1*0406 molecules, was incubated at room temperature for 48 h with DRB 1 *0406 molecules in the presence of nonlaheled peptides tested for DR binding and a protease inhibitor mixture. Inhibitory effects of peptides on the binding of 0405BP3 to DR4 molecules were plotted at various concentrations of the peptide inhibitor, and the concentration of peptide inhibitor yielding 50% inhibition (ICs$ was determined.

Generation and characterization of an Ag-specific T cell clone

Ag-specific T cell lines were established from a donor heterozygous for HDI-DRBI *0401-DQA1*0302-DQBI*0301 and DRBI W406- DQAI*0301-DQBZ *0302 haplotypes, as described (26). T cell cloning was done using Terasaki plates and by limiting dilution at 0.3 T cell/well in the presence of 3 X IO'/well irradiated (3000 cGy) autologous PBMC, prepulsed with the peptide in RPMI 1640 medium (Life Technologies, Grand Island, NY) supplemented with 10% heat-inactivated plasma from human males, antibiotics, L-glutamine, and human rIL-2. Peptide-pulsed PBMC was prepared by incubating PBMC with 40 pM of peptide at 37°C for 2 h, followed by washing twice with the medium. Growing microcultures were then expanded at weekly intervals by feeding with irradiated PBMC pulsed with peptide and complete medium supplemented with human rIL-2. Sur- face markers of the T cell clone were analyzed by immunofluorescence staining using anti-CD3, anti-CD4, anti-CD8, or anti-ap TCR mAbs conjugated with FITC or phycoerythrin (Becton Dickinson, San Jose, CA) and flow cytometry, using a FACScan (Becton Dickinson). Quantitation of IL-4 and IFN-y in supernatants of a T cell done was done as described (26). and as follows: T cell clone (3 X IO4) was cultured in 200 pl of medium in 96-well flat-bottom culture plates in the presence of autologous irradiated PBMC ( I X IO'), prepulsed with a peptide (see below). After 48 h, the supernatants were collected, and stored in aliquots at -80°C until

'Abbreviations used in this paper: ICsu, concentration for 50% inhibition; p, position.

use. To measure IL-4 and IFN-y in the supernatants, Quantikine human IL-4 (R&D Systems, Minneapolis, MN) and IFN-y ELISA kit (Otsuka, Tokyo, Japan) were used according to the manufacturer's instructions and in duplicate for each culture supernatant.

Ag-specific proliferative response of the T cell clone

Ag-specific proliferation of the T cell clone was investigated as described (26). Peptide-pulsed PBMC or L cell transfectants expressing HLA-DR were used as APC, and peptide pulse of L cell transfectants was done as f o h v s : L cell transfectants were treated with 20 pg/ml of mitomycin C (Wako, Osaka, Japan), as described (28), and 3.5 X 10' cells/well were incubated with DMEM medium supplemented with 10% heat-inactivated FCS and 40 p M of peptide in a 96-well plate for 16 h, followed by three washings with HBSS (Nissui Pharmaceutical Co., Tokyo, Japan). L cell transfectants BI8D7 (29) carrying the DRA gene plus the DRBI*0406 gene were distributed at the 1 I th International Histocompatibility Work- shop (30). The T cell clone (3 X lo4) was cultured with irradiated (3000 cGy) APC for 72 h and pulsed with 1 KCi/well of ['HITdR for the last 16 h; then cells were harvested and the incorporated radioactivity was measured by liquid scintillation counting. In some experiments, proliferative responses of YN5-32 to APC prepulsed with several doses of each analogue peptide (0.5, 5 , and SO pM) were compared by calculating percentages of response to the wild-type peptide (percentage of wild-type response). Blocking of the proliferative response was investigated by adding anti-HLA class 11 mAbs HU-4 (anti-HLA-DRBI + DRB5 framework) (31), HU-18 (anti-HLA- DQ3) (32), or B7/21 (anti-HLA-DP monomorphic; American Type Culture Collection (ATCC), Rockville. MD) throughout the culture period.

Observation of TCR antagonism, partial agonism, and induction of T cell anergy mediated by analogue peptides

To determine whether analogue peptides could function as an antagonist for TCR, PBMC (2 X 105/well) were first pulsed with a suboptimal concentration (3 pM) of wild-type M12p54-68 for 2 h at 37°C in a 96-well flat-bottom plate, washed twice with medium, then irradiated (3000 cGy). PBMC prepulsed with peptide were incubated together with three concentrations of each soluble analogue peptide (25, 50, and 200 pM) for 12 h, and T cells (3 X I04/well) were added. Cultures were incubated for 72 h at 3 7 T , and ['HH]TdR uptake by T cells was determined as described above. Antagonistic activities of analogue peptides were expressed as percentage of inhibition of the wild-type response, calculated as follows: (1 - response to the wild-type peptide in the presence of an analogue peptide/ response to the wild-type peptide alone) X 100. Inducibility of anergy in YN5-32 by analogue peptides was investigated as follows: PBMC (8 X 104/we11) were incubated alone or with analogue peptides (100 pM) for 12 to 16 h at 37°C in a 96-well flat-bottom plate, and nonadherent cells were washed three times with medium, then irradiated (3000 cGy). T cells (6 X 104/well) were added to the wells, and the culture was incubated for 48 h at 37°C. T cells were then harvested from the wells and incubated for 72 h at 37°C with irradiated PBMC (2.2 X 10s/well), prepulsed for 2 h with M12pS4-68 (10 pM). Cell proliferation was determined as described above. To observe partial activation of the T cell clone in response to analogue peptides presented by DRB 1 *0406 molecules, the T cell clone was harvested on day 6 or 7 after feeding with irradiated and peptide- pulsed DRB 1*0406-positive PBMC plus human rIL-2. T cells (1 X IO6) were incubated in 300 pl of complete medium containing 100 p M of soluble peptide for presentation of peptides to T cells by HLA-DR expressed on the T cell surface or peptide-prepulsed plastic surface-adherent monocytes for 24 h in 48-well microculture plate, and cells were washed three times. T cells were then incubated with anti-CD3, anti-CD49d, anti- CD44 (PharMingen, San Diego, CA), anti-CD4, anti-CD25. anti-CD28 (Becton Dickinson), anti-CD54, anti-CD69 (Anceli, Bayport, MN), anti- CDl la (LFA-1) (TS2/14, ATCC), or anti-CD95 (Fas) (CHI I ; MBL, Nagoya, Japan) mAbs, followed by staining with FITC-conjugated goat anti-mouse IgG Ah F(ab'), (PharMingen), if Abs were not labeled directly with fluorescence and analyzed by FACScan (Becton Dickinson). Re- sponses of T cells to each analogue were compared by calculating percentage to response to the wild-type peptide as follows: (value obtained with an analogue peptide - that with medium alone)/(values obtained with the wild-type peptide - that with medium alone) X 100.

HLA typing

HLA class I1 (DR, DQ, DP) alleles of PBMC donors and L cell transfectants were determined by investigating hybridization between HDI-DR, DQ, DP genes amplified by PCR and sequence-specific oligonucleotide probes, as described elsewhere (28). The nomenclature of the HLA class II alleles was according to the WHO Nomenclature Committee for factors of the H L 1 system (33).

The Journal of Immunology 3785

Results Specificity of a T cell clone, YN5-32

Freshly isolated PBMC from donor YN responded to several independent peptides among overlapping peptides covering an entire sequence of streptococcal M12 protein. Among these peptides, the M12p50-72 was suggested to be a major T cell epitope in the M12 protein because the number of wells containing T cells specific to this peptide was the largest among overlapping peptides when PBLs (1 X 105/well) were cultured in 96 wells of a microculture plate. Thereafter, a T cell clone YN5-32 was established from the donor YN heterozygous for HLA-DRBl*0401 and DRB 1*0406 by stimulating PBMC with M12p50-72. The core sequence of M12p50-72 was determined by investigating the proliferative response of YN5-32 to peptides truncated at N or C terminus, and M12p54-68 was identified to be a minimum length peptide that stimulated as strong a response of YN5-32 as did M12pS0-72 at several different concentrations (not shown). The residues from 57 to 65 of the M12p54-68 peptide are referred to as p l (position 1) through p9 in this study.

The Ag-presenting HLA class I1 molecule was determined to be the HLA-DRB 1 *0406 complex by 1) inhibition of a proliferative response to the peptide of YN5-32 by mAb HU-4 specific to monomorphic determinants of HLA-DRBl or DRB5 gene prod- ucts, and 2) presentation of the peptide to YN5-32 by allogeneic irradiated PBMC positive for DRB 1*0406 and a mouse L cell transfectant expressing human HLA-DRA plus DRB 1 *0406 genes (data not shown). Flow-cytometric analysis of the clone YN5-32 revealed the helpedinducer T cell phenotype (CD3+, TCR-ap+, CD4+. CD8-). In response to M12p54-68, YN5-32 produced a large amount of IFN-7 (9 ng/ml) and a small amount of IL-4 (420 pg/ml) detected in the supernatant.

Proliferative responses of YN5-32 to a large panel of analogue peptides carrying single residue substitutions at residues 57 (p 1) to 65 (p9) of M 1.2~54-68 peptide

APC was prepared by pulsing PBMC with three concentrations (0.5, 5, and 50 pM) of each peptide, and then cultured with YN5-32 to investigate proliferative responses (full agonism) to the analogues. According to the frequency of analogue peptides stimulating proliferative responses, nine amino acid residues from Leus7 (pl) to Ser65 (p9) were roughly separated into three groups as follows.

As shown in Figure 1, C, H , and I , more than 50% of the analogue peptides tested and carrying substitutions at either Gln59 (p3), Leuh4 (p8), or Serh5 (p9) stimulated various magnitudes of proliferative responses in YN5-32. Thus, six analogues, including Q59N, T, V, L, I, and M (Q59N stands for an analogue peptide carrying replacement of Gln with Asn at residue 59), stimulated large or intermediate proliferative responses in YN5-32, and four additional analogues, including Q59E, S, C, and A, stimulated small responses. Analogues carrying replacements to charged amino acids at Gln59 (p3) frequently abrogated stimulatory activity. Similar observations were made for analogues at either Leuh4 (p8) or Serbs (p9), and the analogue peptides L64D, S65E, and S65D lost full agonistic activity to YN5-32, indicating dominant negative effects of single residue replacements to acidic amino acids on T cell proliferation. Analogues with other substitutions stimulated large or intermediate proliferative responses in YN5-32.

As shown in Figure 1, A, D, F , and G, intermediate number (26-37%) of analogue peptides with substitutions at either Leu57 ( P I ) , Alah0 (p4), A d 2 (p6), or Gluh3 (p7) stimulated significant proliferative responses. Four analogue peptides, including L57V, I, F, or M, stimulated marked proliferative responses of YN5-32. The other three analogues, L57G, L57A, and L57W, stimulated small but definite responses at a peptide concentration of SO pM. Thus,

analogue peptides with hydrophobic amino acids (except Tyr and Pro) at residue L e d 7 (p4) stimulated proliferative responses in YN5-32, with various degrees of magnitude. Conversely, all 10 analogue peptides with hydrophilic amino acids lacked such activity. Five analogue peptides, AOT, S, C, V, and I, among analogue peptides with substitutions at residue Ala-60 (p4) stimulated proliferative response in YN5-32 (Fig. 1 D). In case of the residue Am6’ (p6), analogues, N62D, T, A, V, and I, stimulated intermediate or small proliferative responses in YN5-32 (Fig. 1F). In analogue peptides with substitutions at residue G1d3 (p7), peptides E63H, D, Q, G, A, and M, stimulated proliferative responses in YN5-32 to various degrees (Fig. 1G).

On the other hand, in analogue peptides with substitutions at G1u5* (p2) and Tyr6’ (p5) (Fig. I, B and E), it is clear that a relatively small number (10.5%) stimulated the proliferative response. Thus, analogue peptides E58D, E58Q, Y61F, and Y61M, stimulated the proliferation of YN5-32, a finding that means that the T cell responses are sensitive to substitutions of these two residues of the peptide, as compared with the other seven residues.

TCR antagonism mediated by analogue peptides

As shown in Figure 2, in the presence of a soluble form (50 pM) of an analogue peptide E58A. proliferative response of the T cell clone YN5-32 stimulated with autologous irradiated PBMC, prepulsed with 3 pM of M12pS4-68 for 2 h, was inhibited com- pletely to the level of 1.2% of response to the wild-type peptide alone. This inhibition was not due to competitive inhibition of M12p54-68 binding to DRB 1*0406 molecules by the analogue peptide; it was more likely due to TCR antagonism. Two other irrelevant peptides, AAFAALANAA and CD20p26-45 (GPKPL FRRMSSLVGPTQSFF), with a stronger affinity (IC5() = 0.42 and 3.2 pM, respectively) to DRB1*0406 than E58A (IC5,, = 80 pM), showed no significant inhibitory effects on the response of YN5-32 to M12p54-68, at the same concentration as the E58A peptide, as was the case with another irrelevant Der f 2 p109-126 (KIAPK SENVVVTVKLIGD) peptide with no affinity to DRB 1 *0406 molecules. Thus, even in the presence of one of the strong binders to DR4, the wild-type peptide stimulated strong responses, reaching 97 and 88.5% of the response seen with the wild-type peptide alone. Then, many (9 of 17, 52.9%) non-full agonistic analogue peptides carrying substitutions at Glus8 (p2) exhibited a strong or intermediate TCR antagonism (Fig. 3 8 ) . As shown in Figure 3, C, E, and G, similar observations were made for analogues at Gln59 (p3), Tyr6’ (p5), or Gluh3 (p7), in which 6 (66.7%) of 9, 10 (58.8%) of 17, or 10 (76.9%) of 13 non-full agonistic peptides exhibited strong TCR antagonism, respectively.

Few non-full agonistic analogue peptides exhibited TCR antagonism at residues Leus7 (pl), Alah0 (p4), and A d 2 (p6) (Fig. 3, A , D, and F ) . Thus, none of 12 non-full agonistic analogues at Leus7 (pl), only 2 (14.3%) of 14 non-full agonistic analogues at Alah0 (p4), and only 1 (7.1%) of 14 non-full agonistic analogues at A d 2 (p6) exhibited TCR antagonism for this T cell clone. Similar observations were obtained for two other residues, Leuh4 (p8) and Ser65 (p9), and none of three non-full agonistic analogues, L64D, S65E, and S65D, showed TCR antagonism (data not shown).

Partial activation mediated by some TCR antagonistic analogue peptides

Changes in expression levels of T cell surface markers markedly influenced by recognition of the wild-type peptide were investigated in YN5-32 stimulated with antagonistic analogue peptides. As shown in Figure 4, the TCR antagonistic analogue peptide E63V partially activated YN5-32 to increase cell size and expression levels of CD4, CDl la (LFA-I), CD28, CD49d (VLA-4), and CD95 (Fas). Changes in these six parameters were manifest to the

3786 H U M A N T CELL RESPONSE T O ALTERED PEPTIDE LIGANDS

100

eo AAFAALANAA 60 Glu-58 y1

I (P2) (GPKPLFRRMSSLVGPTQSFF) 3'2 CD20 p26-45

RKHEDQNTSCGAVLIFMWYP M12p54-68 E58A -80 "

(NRDLAQAYNELSGEA)

100

n m

a " Gln-59 v ) " C m

RKHEDWTSCGAVLIFMWYP (P3)

4 " n 2 108

* I (P4)

'Z 100

80

Q m Ala-60

2 RKHEDONTSCGZWLIFMWYP w

LII 860 - * Tyr-61 a m (P5) v) s O RKHEDPNTSCGAVLIFMWEIP 0 "

Glu-63 (P7)

Leu-64 (Pa)

O " RKHEDQNTSCGAV@IFMWYP

m

60 Ser-65 I

O RKHEDPN!CW!GAVLIFMWYP Basic Acidic Neutral Hydrophobic

FIGURE 1. Full agonistic activity of analogue peptides carrying single amino acid substitutions at either Leu" (p l ) ( A ) , CIuS8 (p2) ( B ) , GlnS9 (p3) (C) , Alabo (p4) (D), Tyr6' (p5) ( E ) , AmbZ (p6) ( F ) , G I u ~ ~ (p7) (C) , Led4 (p8) (H ) , or Ser65 (p9) ( I ) of M12p54-68 peptide. Prolifer- ative response of YN5-32 to APC pulsed for 2 h with 0.5 pM (closed bar), 5 pM (hatched bar), or 50 pM (open bar) of each analogue peptide was standardized by calculating the percentage to the response to the APC pulsed with 50 pM of wild-type peptide. Amino acids substituted for are indicated by a one-letter code under each panel, and are clustered in four groups, based on homology in their characters. The amino acid at the indicated residue of the wild-type peptide is boxed. Each assay was done in duplicate cultures, and mean values are given. SE of the mean values did not exceed 20%. One hundred percent wild-type response ranged from 12,000 cpm to 48,000 cpm, and medium control responses without peptides were less than 300 cpm.

FIGURE 2. Inhibition of proliferative response of YN5-32 to the wild-type Ml2p54-68 peptide mediated by TCR antagonism of an analogue peptide E58A. The assay for TCR antagonism was done as described in Materials and Methods. One irrelevant peptide, Der f 2 p109-126 (KIAPKSENVVVTVKLICD), showing no binding to the DRB1*0406 molecules, and two peptides, AAFAALANAA and CD20p26-45 (GPKPLFRRMSSLVCPTQSFF), which bind to the DRBl*O406 molecules with higher affinity than the analogue peptide E58A, were utilized as control peptides. The binding capacity of each peptide to the purified DRBl*O406 molecules was expressed as IC,, values for binding inhibition assay. Proliferative responses of YN5-32 are expressed as percentages of the response to the APC pulsed with wild-type peptide in the absence of analogue peptides. One hundred percent wild-type response was equivalent to 22,872 cpm, and medium alone culture gave 322 cpm. Mean cpm of triplicate response was analyzed, and the SE of mean cpm did not exceed 15%.

extent that they nearly reached the level induced by the wild-type M12p54-68 peptide. Increases in expression levels of CD25 (IL- 2R) and CD44 were small, but definite. The wild-type peptide induced a marked down-modulation of CD3 expression and up- regulation of CD54 and CD69 expressions, whereas none of the analogue peptides with TCR antagonistic activities did so. As shown in Table I, 1 (E58N) of 9 antagonistic analogue peptides at residue Gh5* (p2), 1 (Q59G) of 6 antagonistic analogues at residue GluS9 (p3), 2 (Y61V and W) of 10 antagonistic analogue peptides at residue Tyr6' (p5), and 5 (E63V, S, C, K, and I) of 10 antagonistic analogue peptides at residue GIuh3 (p7) partially activated YN5-32, as described above, at the concentration that did not induce proliferative responses in YN5-32. None of 3 antagonistic analogue peptides, with substitutions at Ala6' (p4) or Ambz (p6), showed partial activation. None of these 9 analogue peptides with TCR antagonism and partial activation or 29 other antagonistic analogue peptides induced anergy in this T cell clone, when either irradiated PBMC, adherent monocytes, or mouse L cell transfectants expressing DRB1*0406 were used as APC (data not shown). As for production of IFN-y by the T cell clone stimulated with the above 9 analogue peptides exhibiting TCR antagonism and partial activation, none stimulated production of IFN-y, whereas the wild- type peptide did stimulate production of a significant amount of IFN-y (data not shown). Figure 5 summarizes the frequencies of three types of T cell responses, TCR full agonism, and TCR antagonism with or without partial activation, in recognition of analogue peptides carrying single residue substitutions at each amino acid residue of the antigenic peptide.

Discussion We analyzed responses of a human T cell clone to a panel of analogue peptides carrying single residue substitutions in an antigenic peptide, the objective being to investigate frequencies and characteristics of altered T cell responses in recognition of altered peptide ligands. At least three different recognition patterns of analogue peptides by TCR were observed. The first, designated as


face of cells, without any cell proliferation, IFN-y production, or anergy induction.

As summarized in Figure 5, according to differences in frequencies of analogues exhibiting the above three effects on TCR, nine amino acid residues from residues numbers 57 (pl) to 65 (p9) were roughly separated into four groups. Group I residues include Leus7 (pl), Ala6’ ( ~ 4 ) . and A d 2 (p6), and analogues at these residues rarely showed TCR antagonism, and 20 to 40% of them showed full agonism. Group I1 residues consist of Glu5* (p2) and Tyr6’ (p5), and group 111 residues include G l d 9 (p3) and G h h 3 ( ~ 7 ) . Analogues at either group I1 or 111 residues frequently showed TCR antagonism, and 30 to 50% of the analogues did so. These two groups differed in terms of frequencies of analogues exhibiting full agonism. Thus, 10.5% of analogues at group I1 residues acted as full agonists, whereas 30 to 50% of the analogues at the group 111 residues did so. Two residues, L e d 4 and Serh5, belong to the group IV residues, and more than 80% of analogues at these residues showed full agonism, except for analogues with substitutions to acidic amino acids; these amino acids had dominant negative effects on TCR full agonism.

We reported earlier on differences in peptide-binding motifs between two closely related DR molecules, DRB1*0405 and DRB 1 YO406 molecules, determined by investigating binding capacity to these DR molecules of many analogue peptides derived from a self peptide bound to DRB 1*0405 molecule (7). The results corresponding to the present study are summarized as follows: 1) three amino acid residues (DR anchors) on the DRB I *0406-binding peptides are important for binding to the DR molecule; 2) the first anchor, closest to the N terminus of the peptides, must be hydrophobic amino acids, except for Gly and Pro; 3) the second anchor, which is three residues apart from the first one in a direc- tion to C terminus of the peptide, is hydrophobic, but some non- basic hydrophilic residues are also capable of binding; and 4) the third anchor, at two residues apart from the second anchor, is promiscuous, but the binding affinity was variable.

Based on all of the data, group I residues of M12p54-68 peptide are suggested to be DR anchor residues. At the putative first DR anchor residue Leu5’ (pl), analogue peptides with hydrophobic amino acid replacements, including Val, Ile, Phe, or Met, stimulated strong proliferative responses of YN5-32, whereas analogues with hydrophilic amino acid replacements did not. These observations fit well characteristics of the first DRB 1 *0406 anchor residue, and suggest that the first DR anchor residue may function mainly as a DR anchor alone, and substitution of this residue might not much influence conformation of the peptide recognized by TCR, at least in the M12p54-68 peptide. The M12p54-68 peptide has double DRB 1 *0406-binding motifs, L57-A60-N62 and Y61-L64-G66, but replacements of Tyr6’ with aliphatic amino acids abrogated full agonistic activity of the peptide; therefore, Tyr6’ may not be the first DR anchor residue. As for the putative second DR anchor residue, Alaho (p4), five analogue peptides with substitutions to hydrophobic or neutral amino acids stimulated the T cell clone, but all other analogues that were thought to be bound to DRB1*0406, as judged by our previous study (7), failed to stimulate the T cell clone. Similar observations were made for analogues at residue Am6* (p6), which may be the third DR anchor residue. To verify these assumptions, the direct binding assay between analogue peptides and DR4 molecules is essential. We set up a binding inhibition assay utilizing Iz5I- or biotin-labeled syn- thetic peptides and affinity-purified DR molecules or DR molecules expressed on mouse L cells transfected with HLA-DR genes. Presumably because of the presence of double DRB 1 *0406-binding motifs L57-A60-N62 and Y61-L64-G66 in the peptide and the weak binding affinity to DR molecules of M12p54-68 and its analogue peptides, it was difficult to precisely identify DR anchor

a i N N N N N N N N 1

Gln-59 9 4 (P3)

0 * ‘S ”

O RKHED@NTSCGAVLIFMWYP

”

Tyr-61 (P5)

i

FIGURE 3. TCR antagonistic activities of analogue peptides derived from M12p54-68 carrying substitutions at either Leu” (p l ) ( A ) , CIu5’ (p2) ( B ) , Gln5’ (133) ( 0 , Ala“” (p4) (DL Tyr6’ (p5) ( 0 , A d 2 (p6) ( 0 , or C I d 3 (p7) (G) residues. As described in Materials and Methods, TCR antagonistic activity of the analogues was expressed by the percentage of inhibitory effect of the analogue peptide on proliferative responses of YN5-32 to the wild-type peptide, which ranged from 10,000 to -15,000 cpm. Three concentrations of analogue peptides, 25 pM (closed bar), 50 pM (hatched bar), and 200 pM (open bar), were used for observation of TCR antagonism. N T means not tested for TCR antagonistic activity, because the analogue peptide showed full agonism. Analogue peptides with substituted amino acids indicated by asterisks partially activated the T cell clone (see below).

TCR full agonism, induced proliferative responses of T cells, to various magnitudes. The second is TCR antagonism to inhibit T cell proliferation to the wild-type peptide by a mechanism other than competitive inhibition of binding to HLA of the wild-type peptide. The third is TCR antagonism with partial activation, in which T cells were partially activated to increase cell size and expression levels of CD4, CD28, CD49d (VLA-4), and CD95 (Fas), as well as that of CD1 la (LFA-I) molecules on the sur-

3788 HUMAN T CELL RESPONSE T O ALTERED PEPTIDE LIGANDS

8 P)

0 v N

8

0

pm 2% 2 0 S O 02 us1

8

E ) E N

0 0 N N

0 0 a

C c g r

3 0 USZ

c

a Ip

0 !n

0 0

4 FL 1H FLZH FL t H

Relative Log Fluorescence (Channel Number) FIGURE 4. Marked increases in cell size and cell surface expression of CD4, CD11 a (LFA-l), CD28, CD49d (VLA-4), and CD95 (Fas), and marginal increases in cell surface expression of CD25 (IL-2Ra) and CD44 in YN5-32 by recognition of an analogue peptide E63V with TCR antagonistic activity. YN5-32 was incubated with the wild-type peptide, E63V, or medium for 24 h, and analyzed by immunofluorescence staining and flow cytometry, using FACScan for cell size (forward scatter of cell) (A), CD4 (B ) , CD1 l a (LFA-1) (C) , CD28 (D), CD49d (VLA-4) (E), CD95 (Fas) ( F ) , CD25 (IL-2Ra) (G), and CD44 (H). Thick dotted lines indicate profiles of cells incubated with the wild-type peptide, and thin dotted lines indicate those incubated with medium alone. Thick solid lines indicate profiles of cells incubated with the analogue peptide E63V. Negative controls for staining, using isotype-matched normal mouse IgG conjugated with fluorescence, are indicated by thin solid lines. Mean values of each parameter in 10,000 cells are given.

Mean fluorescence intensity

Antigenicity Peptide FSCc CD4 C D l l a CD28 CD49d CD95 CD25 CD44 CD54 CD69 CD3 Induction Anergy

Full agonist Wild-type 100 100 100 100 100 100 100 100 100 100 100 M12 p54-68 (180)d (1564) (436) (1 14) (295) (309) (2966) (1298) (142) (47) (60)

No stimulation None 0 0 0 0 0 0 0 0 0 0 0 (94) (750) (210) (53) (190) (155) (185) (512) (67) (10) (161)

E58N 54.9 40.2 70.7 50.4 47.6 85.9 4.6 27.3 0.2 0 -14.8 NO Q59G 56.3 64.2 78.4 93.1 78.1 89.5 6.5 17.1 0

Partially activating Y61 V 2.5 -4.2 NO

46.5 38.1 62.5 47.5 57.1 56.8 4.3 13.2 0 0 -14.8 NO

E63V 4.8 17.1 0.4 0.9 -12.8 NO

51.2 74.6 81.6 122.0 92.3 92.3 6.5 17.3 0.9 1.7 0 E63S

No 47.7 52.3 65.1 75.2 83.8 78.6 5.9 18.4 0.1 1.4 -9.9 NO

E63C 40.6 63.0 65.9 84.0 73.3 75.2 5.6 12.0 0 1.1 -1.9 NO E63K 40.6 25.1 56.8 30.1 47.6 64.1 3.5 18.1 0 E631

0 -13.8 NO 48.8 52.3 47.2 64.2 57.1 64.1 4.7 12.0 0 0 -9.9 NO

antagonist Y61 W 53.5 50.2 71.2 70.0 66.6 77.9

T cell clone YN5-32 was incubated with the 100 p M of peptide, which did not induce a proliferative response in YN5-32, for 24 h (3 h for CD3 down-modulation) and washed three times with medium, followed by immunofluorescence staining and flow cytometry.

'Values are standardized by calculating percentage to response to the wild-type peptide, as described in Mafeerials and Methods. FSC, foward scatter of cell. FSC and mean fluorescence intensity values for response to the wild-type peptide and medium alone are indicated in parentheses. Mean fluorescence intensities

for negative control staining of T cells with fluorescinated normal mouse IgG or goat anti-mouse IgC were <lo.

residues in the M12p.54-68 peptide (data not shown). Therefore, at this point in time, we cannot conclude that the group I residues are DR anchor residues.

Inhibitory effects of non-full agonistic peptides on the proliferative response of YN5-32 to wild-type peptides were most likely due to TCR antagonism (18, 19) mediated by analogue peptides, for two reasons. 1) The analogue peptides exhibited a stronger inhibitory effect than the two irrelevant peptides with a higher affinity to DRB1*0406. There were 20-200-fold differences in

IC,, values between the analogue peptide and these two strong DR-binding peptides. Therefore, it is unlikely that the analogue peptides inhibited the response of YN5-32 by competitive inhihi- tion of the binding of wild-type peptide to the DR molecules. 2) The analogue peptides did not induce anergy in YN5-32.

Characteristic features of the group I1 and 111 residues are as follows: 1) As many as 35 (63.6%) of 55 non-full agonistic analogue peptides with relatively nonconservative amino acid substitutions showed TCR antagonism. Because TCR antagonism itself


O T C R full agonlsm + TCR antagonlsrn -TCR lull agonlsrn

FIGURE 5. Summary of responses of the 1 cell clone YN5-32 to 156 analogue peptides carrying single residue substitutions at residues from 57 ( p l ) to 65 (p9) of M12p54-68 peptide. In the upper panel, percentages of analogue peptides exhibiting TCR full agonisrn are indicated by closed bars, and MI2 @ 8 those of analogue peptides exhibiting either TCR 4-68 full agonisrn or TCR antagonism are indicated by open bars for each residue on M12p54-68 peptide. In the lower panel, percentages of analogue peptides exhibiting TCR antagonism are indicated by open bars, and those of analogue peptides exhibiting TCR antagonism with partial activation are indicated by closed bars, for each residue on M12p54-68 peptide. 0 TCR antagonlsm

-TCR antagonlsm

. . . . . . .....

wlth partlal actlvatlon

proves binding of the analogue peptides to DR molecules, as reported (20), 55 (72.4%) full agonistic or antagonistic analogue peptides among the total of 76 analogues probably bind to DR molecules. 2) Nine (25.7%) of 35 TCR antagonistic analogue peptides with relatively conserved amino acid substitutions partially activated the T cell clone to increase cell size and cell surface expressions of CD4, CDl la (LFA-l), CD28, CD49d (VLA-4), and CD95 (Fas) in the T cell clone. This was the most prominent at Glu6’ (p7), and 5 of 10 antagonistic analogues partially activated the T cell clone. 3) The major difference between group I1 and 111 residues is in frequencies of full agonistic analogues, and group 111 residues provided greater agonistic analogues than did group 11. Only 4 (10.5%) of 38 analogue peptides, with substitutions at group I1 residues Glu5’ (p2) and Tyr6’ (p5) to relatively conserved amino acids, stimulated proliferation of the T cell clone. Thus, group I1 residues may be the major contact sites for TCR, and TCR full agonism is very sensitive to substitutions at these residues. The frequent appearance of TCR antagonistic analogues and the relatively low frequency of TCR agonistic analogues substituted at the TCR contact residues were noted by other investigators (20). On the other hand, as many as 11 (57.9%) of 19 analogue peptides at group 111 residues stimulated proliferative responses. It may be that group I11 G l d 9 (p3) and G l d 3 (p7) residues are transitional ones between the major TCR recognition residues and DR anchor residues, and that substitutions at these residues influence conforma- tional changes of the TCR recognition residues, but little DR-binding capacity of the peptide. Analogue peptides with replacements at G l d 9 (p3) of Arg and Lys showed neither full agonism nor TCR antagonism, possibly because of repulsive ionic interactions between basic residues of the peptide and Arg” and/or Arg72 of the DR P-chain.

All of the tested analogue peptides carrying a substitution at either residue 64 (p8) or 65 (p9), except substitutions to acidic amino acids, stimulated the T cell clone, thereby suggesting the dominant negative effect of acidic amino acid at these residues on the recognition by the T cell clone YN5-32. This observation may be explained by a repulsion between acidic amino acid on the peptide and acidic DRP 57 Asp or possible acidic residue(s) on the CDR3 region of the TCR of YN5-32. We favor the former possi- bility, because we observed no binding to the DQ9 molecule of the peptide with the acidic residue at the possible contact site for Asp at DQ9P57 (34). Similar dominant negative effects of some amino acid residues of a peptide on T cell response were noted in a mouse system (35).

Recent crystallographic analyses of the DRl molecule complexed with influenza hemagglutinin peptide HA306-3 18 (17) clearly elucidated the physical basis of interaction between the DR molecule and peptides, and tandem arrangement of DR anchor residues and possible TCR contact residues. Our present observations of possible tandem arrangement of group I putative DR anchor residues and group I1 putative major TCR recognition residues may prove to have some similarity to tertiary structures of the peptide bound to the DR1 molecule.

Partial activation exhibited by some TCR antagonistic analogue peptides clearly indicated that differences in thresholds and/or signal-transduction pathways may exist for each activation marker, such as increases in cell size and expression level of cell surface markers, cytokine production, DNA synthesis, and subsequent cell division. In terms of changes in expression levels of cell surface markers, they were separated into three groups. The first group represents markers of which the expression level was altered markedly in recognition of partially activating TCR antagonistic analogue peptides. CD4, C D l l a (LFA-l), CD28, CD49d (VLA-4), and CD95 (Fas) belong to this group. In the second group are CD25 and CD44, the expression levels of which were to some extent influenced by partial activation. Expression levels of the third group markers, including CD3, CD54, and CD69, were influenced by full agonism mediated by the wild-type peptide, but not by analogue peptides. No down-modulation of CD3 by partially activating analogue peptides was observed in our system, as was reported by other investigators (36). It is interesting to note that 7 (77.8%) of 9 analogue peptides showing TCR antagonism with partial activation have relatively conservative amino acid substitutions, in terms of hydrophobicity or hydrophilicity of amino acids. On the other hand, 21 (80.8%) of 26 analogue peptides with substitutions at group I1 and I11 residues, showing TCR antagonism without partial activation, carry nonconservative amino acid substitutions (substitutions from hydrophilic Glu to hydrophobic amino acids, and those from hydrophobic Tyr to hydrophilic amino acids). In contrast to these observations, all three antagonistic and non-partially activating analogue peptides at the group I residues had conservative amino acid substitutions.

These observations collectively confirm findings in mice that activation mediated by TCR recognition of the ligand is not an odoff event, and it may be quantitatively or qualitatively different depending on affinity and avidity between TCRs and their ligands (37). In mouse systems, several partially activating altered peptide ligands induced anergy and up-regulation of CD25 (IL-2R), as

3790

well as increases in cell size and expression level of C D l l a (LFA-I) in a Thl clone (21). However, in our present study, we found no analogue peptides with such characteristics. The T cell clone and analogue peptides used in this study may provide a good model to dissect complex systems of signal transduction through TCR in human T cell activation. Another interesting question is whether or not partial agonistic and/or antagonistic analogue peptides used in this study induce partial phosphorylation of CD3 (-chains, leading to the absence of phosphorylation in ZAP70, as noted in mouse T cell clones (38, 39). Our observations suggest that pathogenic T cell responses causing autoimmune diseases or allergy may be down-regulated by analogue peptides derived from T cell epitopes by exerting antagonistic activity for TCR in human T cells. This basic knowledge of human T cell responses to altered peptide ligands is useful to search for non-self peptides that mimic self peptides that may trigger autoimmunity (40).

Acknowledgments We thank Dr. A. Wakisaka (Hokkaido University, Japan) for providing anti-HLA class I1 mAbs, Dr. N. Kashiwagi (Kitasato University, Japan) for the gift of KT13, and M. Ohara for helpful comments.

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

I O .

1 I .

12.

13.

14.

15

16

17.

Germain, R. N., and D. H. Margulies. 1993. The biochemlstry and cell biology of antigen processing and presentation. Annu. Rev. Immunol. 11:403. O’Sullivan, D., J. Sidney, E. Appella, L. Walker, L. Phillips, S. M. Colon, C. Miles, R. W. Chesnut, and A. Sette. 1990. Characterization of peptide binding to four DR haplotypes. J. Immunol. 145:1799. Krieger, J. I., R. W. Karr, H. M. Grey, W. Y. Yu, D. O’Sullivan. L. Batovsky, Z. L. Zheng, S. M. Colon, F. C. A. Gaeta, J. Sidney, M. Albertson, M. F. Del

and antigen define residues critical for peptide-MHC binding and T cell recog- Guercio, R. W. Chesnut, and A. Sette. 1991. Single amino acid changes in DR

nition. J. Immunol. 146:233/. O’Sullivan, D., T. Arrhenius, J. Sidney, M. F. Del Guercil, M. Albertson, M. Wall, C. Oseroff, S. Southwood, S. M. Colon, F. C . A. Gaeta, and A. Sette.

allele: identification of common structural motifs. J. Immunol. 147:2663. 1991. On the interaction of promiscuous antigenic peptides with different DR

Hill, C. M.. J. D. Hayball, A. A. Allison, and J. B. Rothbard. 1991. Conforma-

J. Immunol. 147:189. tional and structural characteristics of peptides bmding to HLA-DR molecules.

Sette, A., J. Sidney, C. Oseroff, M. F. Del Guercio, S. Southwood, T. Arrhenius, M. F. Pwell, S. M. Colon, F. C. A. Gaeta, and H. M. Grey. 1993. HLA DR4w4- binding motifs illustrate the biochemical basis of degeneracy and specificity in peptide-DR interactions. J. Immunol. 151:3163. Matsushita, S., K. Takahashi, M. Motoki, K. Komoriya, S. Ikagawa, and Y. Nishimura. 1994. Allele specificity of structural requirement for peptides bound to HLA-DRB1*0405 and -DRBl*0406 complexes: implication for the HLA-associated susceptibility to methimazole-induced insulin autoimmune syndrome. J. Exp. Med. 1801873. Hammer, J., E. Bono, F. Gallazzi, C. Belunis, Z. Nagy, and F. Sinigaglia. 1994.

tion based on peptide side chain scanning. J. Exp. Med. 180:2353. Precise prediction of major histocompatibility complex class 11-peptide interac-

Nagy, and F. Sinigaglia. 1995. Peptide binding specificity of HLA-DR4 mole- Hammer, J., F. Gallazzi. E. Bono, R. W. Karr, J. Guenot, P. Valsasnini, 2. A.

cules: correlation with rheumatoid arthritis association. J. Exp. Med. 181:1847. Hammer, J . , B. Takacs, and F. Sinigaglia. 1992. Identification of a motif for HLA-DRI binding peptides using MI3 display libraries. J . Exp. Med. 176:1007. Hammer, J. , P. Valsasnini, K. Tolba, D. Bolin, J. Higelin, B. Takacs, and F. Smigaglia. 1993. Promiscuous and allele-specific anchors in HLA-DR-binding peptides. Cell 74:197. Rothbard, J. B., R. Busch, V. Bal, J . Trowsdale, R. 1. Lechler, and J . R. Lamb.

Int. Immunol. 1:487. 1989. Reversal of HLA restriction by a point mutation in an antigenic peptide.

Rothbard, J. B., R. Busch, K. Howland, V. Bal, C. Fenton, W. R. Taylor, and J. R. Lamb. 1989. Structural analysis of a peptide-HLA class I1 complex: identification of critical interactions for its formation and recognition by T cell receptor. Int. Immunol. 1:479. O’Sullivan, D., J . Sidney, M. F. Del Guercio, S. M. Colon, and A. Sette. 1991. Truncation analysis of several DR binding epitopes. J. Immunol. 146:1240. Geluk, A,, K. E. Van Meijgaarden, A. A. Janson, 1. W. Drijfhout, R. H. Meloen, R. R. De Vries, and T. H. Ottenhoff. 1992. Functional analysis of DR17(DR3)- restricted mycobacterial T cell epitopes reveals DR-binding motif and enables the design of allele-specific competitor peptides. J. Immunol. 149:2864. Brown, J . H., T. S. Jardetzky, J . C. Gorga, L. J. Stem, R. G . Urban, J. L. Strominger, and D. C. Wiley. 1993. Three-dimensional structure of the human class 11 histocompatibility antigen HLA-DRI. Nature 364:33.

Strominger, and D. C. Wiley. 1994. Crystal structure of the human class I1 MHC Stem, L. J., J. H. Brown, T. S. Jardetzy, J. C . Gorga, R. G . Urban, J. L.

protein HLA-DRI complexed with an influenza vlrus peptide. Narure 368:215.

H U M A N T CELL RESPONSE T O ALTERED PEPTIDE LIGANDS

18. De Magistris, M. T., J . Alexander, M. Coggeshall, A. Altman, F. C. Gaeta, H. M. Grey. and A. Sette. 1992. Antigen analog-major histocompatibility complexes act as antagonists of the T cell receptor. Cell 68:625.

19. Racioppi. L., F. Ronchese, L. A. Matis, and R. N. Germain. 1993. Peptide-major histocompatibility complex class I1 complexes with mixed agonisdantagonist properties provide evidence for ligand-related dlfferences in T cell receptor-dependent intracellular signaling. J. Exp. Med. 177:1047.

20. Alexander, J., K. Snoke, J. Ruppert, J. Sidner, M. Wall, S. Southwood, C. Oseroff, T. Arrhenius, F. C. A. Gatea, S. M. Colon, H. M. Grey, and A. Sette. 1993. Functional consequences of engagement of the T cell receptor by low affinity ligands. J. Immunol. 1SO:l.

21. Sloan-Lancaster, J., B. D. Evavold, and P. M. Allen. 1993. Induction of T-cell anergy by altered T-cell-receptor ligand on live antigen-presenting cells. Nature 363:156.

22. Sloan-Lancaster. J., B. D. Evavold, and P. M. Allen. 1994. Th2 cell clonal anergy as a consequence of partial activation. J. Exp. Med. 180:1195.

23. Evavold, B. D., and P. M. Allen. 1991. Separation of IL-4 production from Th cell proliferation by an altered T cell receptor ligand. Science 252.1308.

24. Soloway, P., S. Fish, H. Passmore, M. Gefter, R. Coffee, and T. Manser. 1991. Regulation of the immune response to peptide antigens: differential induction of immediate-type hypersensitivity and T cell proliferation due to changes in either peptide structure or major histocompatibility complex haplotype. J . Exp. Med. I 741847.

25. Windhagen, A., C. Scholz, P. Hollsberg, H. Fukaura. A. Sette, and D. A. Hafler. 1995. Modulation of cytokine patterns of human autoreactive T cell clones by a single amino acid substitution of their peptide ligand. Immunity 2.373.

26. Ikagawa, S., S. Matsushita, Y. 2. Chen, T. Ishikawa, and Y. Nishimura. 1996. Single amino acid substitutions on a Japanese cedar pollen allergen (Cry j I ) - derived peptide induced alterations in human T cell responses and T cell receptor antagonism. J. Allergy Clin. Immunol. 97:53.

27. Robbins, J. C., I. G . Spanier, S. I. Jones, W. J. Simpson, and P. P. Cleary. 1987. Streptococcus pyogenes type 12 M protein gene regulation by upstream sequences. J. Bacteriol. 169:5633.

28. Kamikawaji, N., K. Fujisawa, H. Yoshizumi, M. Fukunaga, M. Yasunami, A. Kimura, Y . Nishimura, and T. Sasazuki. 1991. HLA-DQ-restricted CD4+ T cells specific to streptococcal antigen present in low hut not in high responders. J. Immunol. 146:2560.

29. Gregersen, P. K., S. M. Goyert, Q. L. Song, and J. Silver. 1987. Microhetero- geneity of HLA-DR4 haplotypes: DNA sequence analysis of LDKTZ” and LD- “TAS” haplotypes. Hum. Immunol. 19:287.

30. Inoko, H., J. G. Bodmer, 1. M. Heyes, S. Drover, J . Trowsdale, and W. H. Marshall. 1992. Joint report on the transfectandmonoclonal antibody component. In HLA 1991, Vol. I . K. Tsuji, M. Aizawa, and T. Sasazuki, eds. Oxford Science Publications, Oxford, p. 919.

31. Hirayama, K., Y . Nishimura, K. Tsukamoto, and T. Sasazuki. 1986. Functional and molecular analysis of three distinct HLA-DR4 /3-chains responsible for the MLR between HLA-Dw4, Dw15, and DKT;?. J. Immunol. 137:924.

32. Kasahara, M., T. Takenouchi, K. Ogasawara, H. Ikeda, T. Okuyama, N. Ishikawa, J . Moriuchr, A. Wakisawa, I. Kikuchi. W. Aizawa. T. Kaneko, N. Kashiwagi, Y. Nishimura, and T. Sasazuki. 1983. Serologic dissection of

485. HLA-D specificlties by the use of monoclonal antibodies. Immunogenetics 17:

33. Bodmer, J. G., S. G. E. Marsh, E. D. Albert, W. F. Bodmer. B. Dupont, H. A. Erlich, B. Mach, W. R. Mayr, P. Parham, T. Sasazuki, G. M. T. Schreuder, J. L. Strominger, A. Svejgaard, and P. 1. Terasaki. 1994. Nomenclature for factors of the HLA system. Hum. Immunol. 41:l .

34. Oiso, M., T. Nlshi, T. Ishikawa, Y . Nishimura, and S. Matsushita. 1996. Differ- ential binding of peptides substituted at putative C-tenninal anchor residues to HLA-DQ8 and DQ9 differing only at p57. Hum. Immunol. In press.

35. Boehncke, W.-H., T. Takeshita, C. D. Pendleton, R. A. Houghten, S. Sadegh-Nasseri, L. Racioppi, J. A. Berzofsky, and R. N. Germain. 1993. The

peptide-MHC molecule interactions and T cell recognition. J. Immunol. 150:33I. importance of dominant negative effects on amino acid side chain substitution in

36. Valitutti, S . , S . Muller, M. Cella, E. Padovan, and A. Lanzavecchia. 1995. Serial triggering of many T-cell receptors by a few peptide-MHC complexes. Nature 375:148.

37. Evavold, B. D., J. Sloan-Lancaster, and P. M. Allen. 1993. Ticking the TCR: selective T-cell functions stimulated by altered peptide ligands. Immunol. Today 14:602.

38. Sloan-Lancaster, J., A. S. Saw, J. B. Rothbard, and P. M. Allen. 1994. Partial T

T cell anergy. Cell 79:913. cell signaling: altered phospho-% and lack of Zap70 recruitment in APL-induced

39. Madrenas, J., R. L. Wange, J . L. Wang, N. Isakov, L. E. Samelson, and R. N. Germain. 1995. 5 phosphorylation without ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267:515.

40. Wucherpfennig, K. W., and J. L. Strominger. 1995. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein. Cell 80:695.

of t cell clone to a large panel of altered peptide

Documents