The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wither, J. E.
Right arrow Articles by Vukusic, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wither, J. E.
Right arrow Articles by Vukusic, B.
The Journal of Immunology, 1999, 162: 2113-2122.
Copyright © 1999 by The American Association of Immunologists

The Orientation and Nature of the Interaction Between Beef Insulin-Specific TCRs and the Insulin/Class II MHC Complex1

Joan E. Wither2,*,{dagger} and Brian Vukusic*

* The Arthritis Center of Excellence, Toronto Hospital Research Institute, Toronto Hospital-Western Division, and {dagger} Departments of Medicine and Immunology, University of Toronto, Toronto, Ontario, Canada


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Recent crystallographic studies suggest that TCR interact with peptide/class I MHC complexes in a single preferred orientation. Although similar studies have not been performed for class II-restricted TCR, it has been proposed that T cell recognition of peptide/class II complexes has similar orientational restrictions. This study represents a functional approach to systematic analysis of this question. Twenty-one mutant Aßd molecules were produced by alanine scanning mutagenesis and assessed for their ability to present species variants of insulin to a panel of beef insulin-specific T cell hybridomas with limited TCR {alpha}- and/or ß-chain sequence differences. We demonstrate that all beef insulin-specific TCR have the same orientation on the insulin/Ad complex, such that the {alpha}-chain interacts with the carboxyl-terminal region of the Aßd {alpha}-helix, and the ß-chain complementarity-determining region 3 interacts with the carboxyl-terminal portion of the peptide, consistent with that observed for crystallized TCR-peptide/class I complexes. Despite this structural constraint, even TCR that share structural similarity show remarkable heterogeneity in their responses to the panel of MHC mutants. This variability appears to result from conformational changes induced by binding of the TCR to the complex and the exquisite sensitivity of the threshold for T cell activation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Activation of Ag-specific Th cells involves interaction between a clonally restricted TCR and processed peptides of Ag bound to MHC class II molecules on the surface of APC 1 . The specificity of the TCR for both Ag and MHC is localized within the variable portion of the TCR {alpha}- and ß-chains, which are assembled from V, D (for ß only), and J segments during development 2 . TCR diversity is generated by combinatorial assortment of these gene segments, variability in the joining position of the gene segments, and random addition of nucleotides at each junction. Although these mechanisms result in the potential for an estimated 1018 different TCR amino acid sequences, the majority of the diversity (~1015 different sequences) lies within the junctional region 3 . Recent crystallographic studies have demonstrated that interaction between the {alpha}- and ß-chains of the TCR and a peptide/class I MHC complex occurs through three complementarity-determining regions (CDR),3 as predicted by the sequence similarity between TCR and Ig gene V regions 3, 4, 5, 6, 7, 8, 9, 10 . While the CDR3 is composed of the highly variable junctional regions, CDR1 and CDR2 are located within the V gene segments and are consequently much less diverse.

On crystallography of peptide/class I and, more recently, peptide/class II MHC complexes, the peptide is bound in a groove formed by two {alpha}-helixes located above a ß-pleated sheet platform (reviewed in 11 . Peptides bound to the class II MHC molecule are generally 14–18 amino acids long and appear to adopt an extended polyproline type II conformation, such that only a few regularly spaced amino acid side chains are directed toward the TCR 12 . The majority of the residual side chains protrude into pockets lined with MHC residues, which have been shown to determine the specificity of peptide binding to the MHC molecule 13, 14, 15, 16 . Many of these pockets are embedded deep within the peptide/MHC complex, so they are not accessible for interaction with the TCR. Consequently, the surface of the peptide/class II MHC complex that is exposed to the TCR consists of both MHC and peptide residues, at least some of which have been conformationally altered by interaction with each other 17, 18, 19, 20 .

Based upon the structural considerations outlined above, several groups independently hypothesized that the TCR was oriented on the peptide/MHC complex such that the highly variable CDR3s interact predominantly with peptide, while the less variable CDR1 and CDR2 interact with residues from the {alpha}-helixes of the MHC molecule 3, 9, 10 . Subsequently, we and others have provided evidence that supports the role of the CDR3s in determining peptide specificity 21, 22, 23, 24 . Nevertheless, it has remained unclear whether the TCR {alpha}- and ß-chain CDR1 and CDR2 interact with MHC in a single preferred orientation.

Several recent crystallographic studies of different TCR-peptide/class I MHC complexes suggest that all class I-restricted TCR may indeed be similarly oriented on the peptide/class I complex, albeit in a slightly different orientation than originally hypothesized 4, 5, 6, 7, 8 . In these studies the TCR was oriented diagonally, such that the TCR {alpha}-chain CDR1 was located over the amino terminus of the peptide, while the TCR ß-chain CDR3 interacted with the carboxyl terminus of the peptide. The general applicability of this orientation for class I-restricted recognition is supported by two additional observations. First, 59 Kb-specific CTL clones share a common response pattern to mutant H-2Kb molecules that is consistent with this orientation 25 . Second, OVA peptides modified at the carboxyl terminus select the ß-chain CDR3 of OVA-specific H-2Kb-restricted TCR as predicted 26 . Although there are no corresponding crystal structures of TCR-peptide/class II MHC complexes, it has been conjectured that this diagonal orientation would permit a better fit between the TCR and peptide-class II MHC complex, and consequently may apply to both class I and class II MHC molecules 5 .

In this study we use a well-characterized panel of beef insulin (BI)/Ad-restricted T cell hybridomas together with 21 Ad ß-chain mutants and four species variants of insulin to explore the nature of interactions between the TCR and insulin/Ad complex. The panel of naturally arising T cell hybridomas, which was derived from immunizations of BALB/c, (BALB/c x BALB.K)F1, or (BALB/c x A/J)F1 mice, all recognize the immunodominant A1–14 peptide derived from BI 21, 27 . While the hybridomas express a variety of different TCR, within the panel there are groups of T cells with identical or similar TCR {alpha}- and/or ß-chains. Comparison of the TCR expressed by the T cell hybridomas with their ability to respond to the various insulins presented by the panel of mutant Ad molecules allowed us to orient the TCR on the insulin/Ad complex. We provide evidence that the hybridomas share a common orientation consistent with that observed for crystallized TCR-peptide/class I complexes, supporting the hypothesis that all TCR share a similar orientation on the peptide/MHC complex. Despite this structural constraint, we demonstrate that the consequence of individual interactions between the TCR and the peptide or MHC molecule is remarkably variable. This variability appears to result from the affect of conformational changes in the TCR and/or peptide on T cell recognition as well as the exquisite sensitivity of the threshold for T cell activation. We discuss the potential implications of these findings for the processes of thymic selection and foreign Ag recognition.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cells

Full-length Aßd cDNA, provided by Dr. R. Germain, was cloned into the EcoRI site of M13 and subjected to site-directed mutagenesis by the method of Kunkel et al. (Muta-Gene Kit, Bio-Rad, Richmond, CA). Mutant plasmids were initially screened by DNA sequencing of the ß1 domain and were subsequently sequenced in their entirety to ensure that no additional mutations or deletions had been introduced. EcoRI fragments containing the mutant cDNAs were then cloned into a eukaryotic expression vector, pR1aneo. This expression vector, created in our laboratory, was produced from pSV2 neo by ablating the EcoRI site and cloning an SV40 promoter, a polylinker containing an EcoRI site, and a polyadenylation signal into the BamHI site. The resultant plasmids were transfected by electroporation into the mutant B cell lymphoma, M12.C3 28 . Because this cell line lacks an RNA transcript for Aßd, transfection with a mutant ß-chain construct rescues I-A cell surface expression. After selection with G418 (Life Technologies, Grand Island, NY), resistant cells were screened for I-A expression by staining with an anti-A{alpha}d mAb, K24-199 29 . Positively staining cells were then subcloned and maintained in medium containing G418. The amount of I-A expressed was quantitated for each mutant using the Quickcal function on a FACScan instrument (Becton Dickinson, Mountain View, CA). Stable transfectants for each mutant were selected such that surface expression of I-A varied by less than twofold.

Insulins and peptides

T cell hybridomas were derived from mice immunized with monocomponent BI (Novo Pharmaceuticals, Copenhagen, Denmark). Insulins used in Ag presentation assays were BI, pork (PI), sheep (SI), and equine (EI; Sigma, St. Louis, MO). The A1–14 peptide of BI was synthesized by the Alberta Peptide Institute (University of Alberta, Edmonton, Canada).

T cell hybridomas

All hybridomas examined were BI/Ad specific and derived by fusion of BW5147{alpha}-ß- with BI-primed lymph node cells. FBD37, FBD45, FBD55, FBD65, FBD84, GBD11, and GBD33 were obtained by immunizing BALB/c mice; BCK37, BCK51, and BCK113 were obtained by immunizing (BALB/c x BALB.K)F1 mice; and H200.1.36 and H200.1.45 were obtained by immunizing (BALB/c x A/J)F1 mice. The derivation, antigenic specificity, MHC restriction, and TCR gene usage of these hybridomas were previously described 21, 27 .

Ag presentation assays

Ag presentation assays were performed using live or fixed M12.C3 I-A ß-chain transfectants as APC. For assays using live cells, 1 x 105 T cells and 1 x 105 APC were cocultured for 24 h at 37°C in wells containing various concentrations of insulin. IL-2 production was assayed by addition of 50 µl of cell culture supernatant to 1 x 104 CTL.L cells. All assays were performed in triplicate, and the results are expressed as the arithmetic mean of the counts harvested following [3H]thymidine incorporation by CTL.L cells. In general, the SD was <10% of the mean. Experiments using fixed cells were performed as previously described 30, 31 . Briefly, fixed APC were prepared by incubation in PBS containing 0.05% glutaraldehyde for 30 s at room temperature followed by addition of an equal volume of 0.2 M L-lysine. The cells were washed extensively and resuspended in PBS or citrate buffer (pH 5.5) containing 2 mM DTT, as a reducing agent, together with various concentrations of peptide or insulin, respectively. Following incubation at 37°C for 18 h the cells were washed with complete medium, and Ag-pulsed APC were cocultured with T cell hybrids as described above.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Generation and characterization of APC-expressing mutant Ad molecules

Table IGo shows the panel of 21 Aßd mutants that were generated in this study. Our approach was to use the technique of alanine scanning mutagenesis to systematically alter amino acids in the {alpha}-helical region of the Ad ß-chain that borders the peptide binding groove. We initially focussed on residues that were predicted by Brown et al. 32 , in their theoretical model of the class II molecule, to point directly toward the TCR or into the peptide binding groove. The recent publication of crystal structures for two different peptide/Ad complexes allows more accurate positioning of the mutated residues in the Ad molecule 20 . As shown in Table IGo, of the 21 mutants, eight point into the peptide binding groove, six point up toward the TCR, four are positioned such that they can interact with both peptide and TCR, and three initially predicted to point up toward the TCR point away from the peptide binding groove such that they are unlikely to interact directly with either the peptide or TCR.


View this table:
[in this window]
[in a new window]
  		Table I. MHC class II Aßd mutants1

 
Expression vectors encoding wild-type or mutant Ad molecules were transfected into the murine mutant B cell lymphoma, M12.C3. This cell line expresses the MHC class I molecules, Kd and Dd, but lacks class II molecules on its surface due to the absence of an RNA transcript for Aßd and an inability to transport Ed to the cell surface 28, 33 . Consequently, transfection with wild-type or mutant Aßd constructs restores I-A, but not I-E, cell surface expression. Using an anti-A{alpha}d mAb, K24-199 29 , surface expression of wild-type and mutant Ad molecules was quantitated in the transfectants by flow cytometry. Subclones whose levels of Ad differed by less than twofold from wild-type levels were selected for further analysis. The only exceptions were N82A and G85A, which expressed Ad suboptimally at levels approximately threefold reduced relative to those of wild-type transfectants. While G85A presents insulin efficiently to all the hybridomas despite its reduced level expression, none of the hybridomas responded to N82A. It has previously been shown that mutation of this asparagine to serine, as seen for Ed in M12.C3, affects intracellular transport and cell surface expression of murine class II molecules 33 . On crystallography this residue plays a critical role in anchoring peptides through hydrogen bonding, suggesting that the failure of expression and presentation observed with mutation of this residue may be related to an inability to efficiently bind peptides 15, 20 .

T cell recognition is dominated by TCR-peptide contacts

Fig. 1Go shows the primary TCR structure of the 17 BI-specific T cell hybridomas used in this study. All the hybridomas recognize the immunodominant A1–14 peptide of BI associated with Ad 21, 27 . The hybridomas that were selected for this study share the following features: 1) they recognize the A chain of BI presented by fixed M12.C3 APC; 2) their TCR {alpha}- and/or ß-chains have been fully sequenced; and 3) each TCR shares structural features with at least one other hybridoma.



View larger version (36K):
[in this window]
[in a new window]
  		FIGURE 1. Amino acid sequences of the TCR CDR of the BI/Ad-restricted T cell hybridomas. A, TCR {alpha} (V-J) and ß (V-D-J) junctional amino acid sequences. The numbering and sequence of the V{alpha}3 gene family members have been previously published (21, 27). For the V{alpha}8 gene family the sequence of V{alpha}8.1 is identical with that of P71{alpha} (34). The sequences of V{alpha}8.2 and V{alpha}8.3 have not been previously published. Numbering of J gene segments is as proposed by Koop et al. (35). Gaps have been inserted before the J gene segment to allow alignment of V and J gene residues. B, CDR1 and CDR2 amino acid sequences of TCR V gene segments with structural similarity to the crystallized 2C TCR. All members of the V{alpha}3 and V{alpha}8 gene families belong to Kabat subgroup I and therefore share framework residues that determine CDR conformation (36).

 
We began by assessing the ability of the hybridomas to respond to BI presented by wild-type and mutant Ad molecules over a range of concentrations. Fifteen hybridomas were subjected to this complete analysis, the results of which are summarized schematically in Fig. 2Go. In general, the ability of mutant MHC molecules to support T cell recognition varied for each BI-specific T cell hybridoma, with anywhere from five to nine (mean, 7.1) mutations leading to marked loss of reactivity.



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 2. Schematic representation of the Ad molecule. Mutated MHC residues that point into the peptide binding groove (A) or up toward the TCR and/or away from the peptide binding groove (B) are numbered. Amino acid residues that when mutated resulted in significant abrogation (>100-fold decrease) of T cell recognition for all (solid filled circles), at least one (striped circles), or none (open circles) of the BI-specific hybridomas are indicated.

 
Of the 21 Ad mutants, only two (E74A and N82A) failed to present BI as efficiently as the wild-type molecule to at least one hybridoma (Fig. 2Go). While N82A is poorly expressed (as outlined above), E74A is expressed at levels comparable to those in mutants that support BI presentation. This suggests that residue 74, which is centrally placed and points into the peptide binding groove, plays a critical role in binding BI and is compatible with the two published peptide-I-Ad crystal structures showing this residue forming part of the pocket that binds residue P4 of the antigenic peptide 20 . Mutation of seven additional residues distributed along the length of the peptide binding groove led to abrogation or marked loss (>100-fold) of T cell recognition for at least one hybridoma. This indicates that the peptide adopts an extended conformation making extensive contacts with the MHC molecule and is consistent with the proposed polyproline structure for class II bound peptides 12 .

Mutation of four of five residues predicted to point directly toward the TCR resulted in complete or marked loss of reactivity for at least one hybridoma, indicating a role for these interactions in T cell recognition. Nevertheless, for each BI-specific T cell hybridoma, the majority of MHC mutants that yielded marked loss of BI reactivity point into the peptide binding groove (mean, 6.1; range, 3–8) rather than directly toward the TCR (mean, 0.9; range, 0–2), suggesting that T cell recognition of BI is dominated by peptide-TCR contacts.

The {alpha}-chain variable regions of BI-specific TCR are located over the carboxyl-terminal portion of the MHC ß-chain

We next sought to determine whether BI-specific T cells expressing the same or similar TCR {alpha}- or ß-chains interact in the same way with the peptide/MHC complex and, if so, whether this would allow us to establish the orientation of BI-specific TCR on the peptide-MHC complex. We began by examining the response of the BI-specific hybridomas to Ad mutations that point directly toward the TCR and do not interact with peptide. We reasoned that if BI-specific T cells with similar TCRs share a common orientation, they would respond similarly to these mutants. The results of this analysis for hybridomas expressing V{alpha}3 gene segments are shown in Fig. 3GoA. All seven V{alpha}3.4-expressing hybridomas, but none of the other hybridomas tested, failed to respond to BI presented by A73V (p = 0.00016). Five of the hybridomas using V{alpha}3.4 were paired with J{alpha}41 and showed similar V-J junctional motifs. Comparison of the abilities of these five hybridomas to respond to the panel of mutants revealed that they shared the same pattern of responses to residues at the carboxyl-terminal region of the ß-chain, defined by an absent response to A73V and increased responses to T77A. In contrast, there was considerable variability in the ability of the hybridomas to respond to E66A. Together these data suggest that the {alpha}-chain of the TCR is located over the carboxyl-terminal region of the MHC ß-chain. This idea is further supported by analysis of the V{alpha}8-expressing hybridomas (Fig. 3GoB). FBD65 and GBD11, which share identical {alpha}-chains but have different ß-chains, have the same pattern of responses to upward-pointing carboxyl-terminal MHC ß-chain mutants 69–84, but differ in their ability to respond to E66A. In contrast, FBD55, which has an identical ß-chain to FBD 65 but expresses a different {alpha}-chain, has a reduced ability to respond to BI presented by T77A. However, these hybridomas respond comparably to BI presented by E66A.



View larger version (41K):
[in this window]
[in a new window]
  		FIGURE 3. BI-specific responses of the hybridomas to Ad mutants that point up toward the TCR. Results for V{alpha}3-expressing (A) and V{alpha}8-expressing (B) hybridomas are shown. Each box contains the results for the hybridoma named in the top left corner, with TCR V{alpha} and J{alpha} gene segment usage shown in brackets. Bars denote the percentage of maximal CTL.L proliferation for increasing concentrations of BI moving from left to right (0, 0.03, 0.3, 3, 30, and 300 µg/ml). Maximal CTL.L proliferation is defined as that induced by supernatants from hybridomas cells stimulated by 300 µg/ml BI presented by the M12C3 wild-type transfectant. Results are representative of at least two determinations per hybridoma.

 
To further define the nature of interactions between the TCR {alpha}-chain and Ad we compared the sequence of the TCR {alpha}-chain CDRs with the pattern of response to the MHC mutants. Precise determination of the CDRs in our hybridomas is facilitated by crystallization of the 2C TCR complexed with the ligand, dEV8/Kb 4, 6 . This TCR (V{alpha}3.1-J{alpha}58; Vß8.2-Dß2-Jß2.4) is structurally similar to many of the TCR examined in this study. Notably, all members of the V{alpha}3 and V{alpha}8 gene families belong to Kabat subgroup I and therefore share framework residues that determine the conformation of the CDRs 36 . This, together with the frequent use of Vß8.3, and to a lessor extent Vß8.2, expressing TCR by BI-specific hybridomas, suggests that the three-dimensional structure of most BI-specific TCRs closely approximates that of 2C. As shown in Fig. 1GoB, the CDRs of V{alpha}3.4 differ from V{alpha}3.5 or V{alpha}3.1 by two amino acids which are located in CDR1 21, 27 . Given the strict association between loss of reactivity to BI with mutation A73V and these sequence differences in CDR1, we propose that CDR1{alpha} is positioned over the Ad ß-chain such that it can interact with MHC residues 73 and 77. This is consistent with the observation that H200.1.36 and H200.1.45, whose {alpha}-chains differ only at CDR1 (Fig. 1Go), differ in their ability to recognize residue T77A (Fig. 3GoB).

The CDR3ß of Vß8.3-expressing BI-specific TCR is located over peptide adjacent to residue 61 at the amino terminus of the MHC ß-chain

Alignment of the TCR {alpha}-chain variable regions of BI-specific hybridomas over the carboxyl-terminal portion of the MHC ß-chain should position their TCR ß-chains over the other end of the peptide binding groove. Confirmatory evidence of this was sought through several different experimental approaches. First, as outlined above, we examined the response of the hybridomas to BI presented by amino-terminal ß-chain MHC mutants that point up (P65A) or away from the peptide binding groove (E59A, S63A). For all hybridomas, regardless of their TCR ß-chain sequence, the response to BI presented by these mutant MHC molecules was comparable to the wild-type response. We next examined the ability of the hybridomas to respond to BI presented by MHC mutants that point into the peptide binding groove (Fig. 4Go). Again, with the exception of D57A, MHC mutants located at the amino-terminal region of the ß-chain (Y60A, W61A, Q64A) presented BI efficiently to the majority of hybridomas, preventing correlation between patterns of reactivity and TCR gene usage. Our final approach was to examine the ability of these mutations to present other species variants of insulin to the panel of BI-specific hybridomas.



View larger version (44K):
[in this window]
[in a new window]
  		FIGURE 4. Affect of Ad mutations that point into the Ag binding groove on BI presentation to the panel of hybridomas. The ß-chain amino acid residues mutated are listed across the top. Hybridomas with TCR {alpha}-chain sequence similarity are grouped together: V{alpha}3.4-J{alpha}41-expressing hybridomas (A), other V{alpha}3-expressing hybridomas (B), and V{alpha}8-expressing hybridomas (C). The results are represented as follows: solid filled boxes, the mutation failed to induce IL-2 secretion by the hybridoma at any Ag concentration tested; cross-hatched boxes, a >100-fold decrease in reactivity to BI; striped boxes, a 20- to 100-fold decrease in reactivity to BI; open boxes, reactivity to BI approached wild-type presentation (<20-fold decrease). Results are representative of at least two independent determinations.

 
The amino acid sequences of SI, PI, and EI insulin differ from BI by one, two, or three amino acids, respectively (Table IIGo). These differences are localized within a region of the A chain (A8–10) that is bounded by an intrachain disulfide bond (A6–11). We have previously shown that BI-specific hybridomas discriminate between these species variants of insulin with the majority of hybridomas having a BI > SI > EI pattern of reactivity and little or no response to PI 21, 27 . Indeed, although several hybridomas showed weak reactivity to PI presented by LB B cell lymphoma cells (decreased 1000- to 3000-fold relative to BI), which present insulin more efficiently than Ad-transfected M12C3 cells, only H200.1.36 responded to PI presented by Ad-transfected M12C3 cells.


View this table:
[in this window]
[in a new window]
  		Table II. Sequence differences between the A1–14 peptide of insulin species variants

 
The pattern of reactivity to the four species variants of insulin was examined with the entire panel of MHC mutants for 10 hybridomas. Representative results from one of the hybridomas, FBD37, are shown in Fig. 5Go. In general, the pattern of reactivity to species variants of insulin presented by mutant MHC molecules was identical with the wild-type pattern. However, mutations T77A and W61A led to acquisition of PI reactivity for some of the hybridomas.



View larger version (39K):
[in this window]
[in a new window]
  		FIGURE 5. Responses of the BI/Ad-restricted hybridoma, FBD37, to four different species variants of insulin presented by the panel of ß-chain mutants. Mutations are listed across the bottom. Bars denote the percentage of maximal CTL.L proliferation for increasing concentrations of each insulin moving from left to right (0, 30, and 300 µg/ml) for each mutation. Maximal CTL.L proliferation was defined as that induced by supernatants from hybridoma cells stimulated by 300 µg/ml BI presented by the M12C3 wild-type transfectant. The results are representative of two independent determinations.

 
Two hybridomas responded to PI presented by mutation T77A. PI responsiveness in these hybridomas did not correlate with a particular TCR {alpha}- or ß-chain. In contrast, initial experiments suggested that the capacity of T cells to respond to PI presented by W61A correlated with expression of Vß8.3-Jß1.1. Subsequent analysis, shown in Fig. 6Go, confirmed that only hybridomas expressing both Vß8.3 and Jß1.1 (n = 6), and none of the other hybridomas tested (n = 8), responded to PI presented by W61A (p = 0.002). Acquisition of reactivity to PI for these hybridomas did not correlate with previous demonstration of weak reactivity to PI by live LB lymphoma cells, nor did it reflect increased binding of PI to W61A because the response of H200.1.36 to PI was not similarly increased.



View larger version (25K):
[in this window]
[in a new window]
  		FIGURE 6. Responses of representative BI-specific hybridomas to BI and PI presented by wild-type Ad or mutated at position 61 of the ß-chain. Each box shows the results for a different hybridoma. Results are expressed as counts per minute of [3H]TdR incorporation by CTL.L cells. Each bar represents the results for a different insulin concentration (0, 3, 30, or 300 µg/ml, increasing from left to right). Asterisks denote statistical significance above background CTL.L proliferation (0 µg/ml). Results are representative of at least two independent determinations.

 
The crystal structure of Ad shows that W61 is positioned such that it interacts with the main chain of residue P8 as well as side chains at P7 and P9 20 . It is likely that Val3 in the A1–14 peptide corresponds to P1 based upon the following considerations: 1) truncation of the A1–14 peptide to A3–14 results in a markedly decreased ability to stimulate T cells, with further truncation leading to complete loss of reactivity 31 ; 2) alkylation of Cys7 (P5) does not affect binding of the A1–14 peptide to Ad, while alkylation of Cys6 (P4) and Cys11 (P9) abrogates binding 30 , consistent with the role of P4 and P9 as anchor residues 20 ; 3) a free thiol at Cys7 (P5) is required for T cell recognition by some BI-specific Ad-restricted hybridomas 30 ; 4) Val at P1, Cys at P4, and Cys at P9 are compatible with the requirement for small hydrophobic residues at P4 and P9 positions to bind to Ad. If Val3 is P1, then Val10 (P8), which is replaced by Ile in PI, is located adjacent to W61ß, suggesting that mutation of this residue leads to local conformational changes in the bound PI peptide, permitting recognition by Vß8.3-Jß1.1-expressing hybridomas.

Despite a common orientation, TCR that share structural similarity show remarkable heterogeneity in their responses to the panel of MHC mutants

As shown in Fig. 4Go, the ability of the hybridomas to respond to BI presented by mutations located in the center of the MHC molecule (ß66–71), which on crystallographic studies are adjacent to P5 and P7, did not correlate with TCR gene usage. For example, comparison of the V{alpha}3.3-J{alpha}41-expressing hybridomas revealed marked variability in the pattern of response to mutations E66, I67, R70, and T71, while comparison of Vß8.3-Jß1.1-expressing hybridomas revealed variability in the response to E66 and T71. Loss of reactivity to BI for these mutations did not correlate with junctional differences, the sequence of the other TCR chain, or the concentration of BI required to induce a maximal response with wild-type Ad. This lack of correlation suggests that there is flexibility in how a given TCR chain can interact with the peptide/MHC complex and suggests that the way one TCR chain contacts the complex may be modified by interactions between the other TCR chain and the complex.

In general, the responses of hybridomas sharing similar TCR {alpha}- or ß-chains to mutations at either end of the peptide binding groove showed less variability than mutations at the center of the groove (Figs. 3Go and 4Go). For example, {alpha}-chain usage had little impact on the ability of hybridomas sharing Vß8.3-Jß1.1 TCR to respond to mutations in the amino-terminal region of the Ad {alpha}-helix (ß57–65). Similarly, hybridomas expressing the same or similar {alpha}-chains showed the same pattern of response to carboxyl-terminal mutations ß73–86 (Figs. 3Go and 4Go), with the exception of A78V for the five V{alpha}3.4-J{alpha}41-expressing hybridomas and H81A for the two V{alpha}8-expressing hybridomas, FBD65 and GBD11 (Fig. 4Go). Further investigation suggested that the variable loss of reactivity to BI presented by A78V or H81A for V{alpha}3.4-J{alpha}41- or V{alpha}8-expressing hybridomas, respectively, resulted from differences between the hybridomas in their reactivities to BI. For the three V{alpha}3.4-J{alpha}41-expressing hybridomas that failed to respond to BI presented by A78V, loss of BI reactivity correlated with a requirement for increased concentrations of BI to produce a maximal response (see responses for wild-type and mutant MHC molecules in Fig. 3Go). Similarly, GBD11, which fails to respond to BI presented by H81A, has slightly decreased reactivity to BI compared with FBD65 (Fig. 3Go). While the magnitude of this reduced reactivity seems insufficient to result in the absence of BI presentation for H81A, analysis of the responses of these to hybridomas to A1–14 peptide and SI is consistent with the interpretation that the loss of BI reactivity is due to this minor reactivity difference. As shown in Fig. 7Go, both FBD65 and GBD11 show a decreased response to truncated A1–14 BI peptide presented by wild-type Ad compared with BI. This results in an almost complete loss of reactivity to this peptide presented by H81A for FBD65. Conversely, GBD11, which demonstrates a heteroclitic response to SI with wild-type Ad (less than a threefold increase in reactivity to SI) 21 (our unpublished observations), acquires reactivity to H81A when it is complexed with this insulin.



View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 7. Variability in the T cell response to the Ad mutant H81A depending upon the Ag presented. Shown are results for FBD65 and GBD11. Results are expressed as the percentage of maximal CTL.L proliferation, with maximal proliferation defined as that induced by supernatants from hybridoma cells stimulated with 300 µg/ml BI presented by the M12C3 wild-type transfectant. The results for BI, A1–14 BI peptide, and SI are shown. Each bar represents the results for a different Ag concentration (0, 3, 30, or 300 µg/ml for BI and SI; 0, 0.6, 6, or 60 µM for A1–14 peptide; increasing from left to right).

 
These data demonstrate that heterogeneity in the response to the panel of MHC mutants for structurally similar TCR arises through two distinct mechanisms. First, the precise manner in which a given TCR {alpha}- or ß-chain contacts the peptide/MHC complex may be modified by interactions between the other TCR chain and the complex. This type of variability appears to occur predominantly in the center of the peptide/MHC complex. Second, minor differences between the hybridomas in the concentration of insulin required to induce a response can lead to altered patterns of reactivity to the panel of MHC mutants, which apparently do not result from differences in TCR specificity. Although our data do not allow us to determine whether these reactivity differences reflect differences in TCR affinity, accessory molecule expression, or downstream events following TCR engagement, they clearly indicate that the T cell activation threshold is very sensitive to minor changes in the peptide/MHC complex.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study we provide evidence that BI-specific TCR share a common orientation on the peptide/MHC complex, with the TCR {alpha}-chain located over the carboxyl-terminal region of the Aßd {alpha}-helix and the TCR ß-chain CDR3 located over the carboxyl-terminal portion of the peptide adjacent to the amino terminus of the MHC molecule. Although the nature of our data does not permit determination of the precise angle of rotation of the TCR on the complex, several aspects of our study indicate that the interaction between BI-specific TCR, which are structurally similar to the crystallized murine TCR 2C, and the BI/Ad complex is similar to that between the 2C TCR and dEV8/Kb 4, 6 . First, BI-specific TCR require both Vß8.3 and Jß1.1 for acquisition of PI reactivity following mutation of 61ß (Fig. 6Go). In the 2C complex, this residue is located at the junction between CDR1ß and CDR3ß. Second, for BI-specific T cells both TCR {alpha}- and ß-chains determine the ability of mutations at the center of the peptide/MHC complex (adjacent to P5) to present BI (Figs. 3Go and 4Go). This is precisely where both {alpha}- and ß-chains interact with peptide and MHC in the 2C complex. Third, mutation of MHC ß-chain residues 59, 63, and 65 has little effect on presentation of BI to the panel of hybridomas, consistent with the fact that little if any of the 2C TCR ß-chain is located over the corresponding residues on Kb (Fig. 3Go and data not shown).

On crystallography of the 2C/dEV8/Kb complex, both CDR1{alpha} and CDR2{alpha} are close to A158, which corresponds to A73 in the Ad molecule, but only residue 51 from CDR2{alpha} directly contacts A158 6, 37 . The observation that there is a strict association between loss of reactivity to BI in the context of A73V and sequence differences in the TCR CDR1{alpha} suggests that these residues may interact directly with A73V (Fig. 3Go). This may indicate that BI-specific TCRs are rotated slightly more counterclockwise on the BI/Ad complex than the 2C TCR on the dEV8/Kb complex. However, the crystal structure of the Ad molecule suggests an alternate explanation. The Ad ß-chain is displaced inward relative to other MHC class II molecules, resulting in a narrower peptide binding groove 20 . This inward displacement may position residue A73V closer to CDR1, allowing interactions to take place.

The data do not support the model of T cell recognition originally proposed by Davis and Bjorkman 3 . In this model the TCR is oriented so that the {alpha} and ß CDR3 span the peptide binding groove while the TCR {alpha}-chain CDR1 and CDR2 are located over the MHC ß-chain. This would position the carboxyl terminus of the antigenic peptide (adjacent to 61ß) close to CDR2{alpha} or leave it bare. Our demonstration that the CDR1ß and CDR3ß of BI-specific TCR interact with this region is not compatible with this model.

Evidence suggesting a TCR orientation on the Ag/class II MHC complex similar to that proposed here has been obtained for one other Ag, conalbumin (CA). Mice carrying TCR {alpha}- or ß-chain transgenes, derived from the CA/Ak-specific T cell line D10, were immunized with wild-type or mutated CA peptides 24 . Peptides mutated at P2 were found to select the sequence of CDR1 or 2{alpha}, those mutated at P5 select CDR3{alpha} and CDR3ß, and those mutated at P8 (which is adjacent to W61ß) select CDR3ß. In a second study, D10 TCR mutants were assayed for their ability to respond to allogeneic or exon-shuffled Ak MHC molecules together with various Ags and peptides 38 . Evidence is provided suggesting that TCR CDR1{alpha} residue 30 interacts with I-A{alpha} residues 52–55, CDR2{alpha} residue 51 interacts with P2, and ß-chain usage affects recognition of I-Aß residue 66. Modelling of the D10 TCR based upon these interactions resulted in an orientation for binding to CA/Ak that was rotated slightly clockwise (as opposed to counterclockwise for the data reported herein) with respect to binding of the 2C TCR to its peptide/class I MHC ligands and consequently to that proposed for BI-specific TCR. Similar minor orientational differences have been described for class I-restricted TCR, indicating that despite a common orientation a limited amount of rotational flexibility is permitted 8 . These data together with evidence from other class II-restricted Ag systems that the amino terminus of the peptide interacts with the TCR {alpha}-chain while the carboxyl terminus interacts with the ß-chain (cytochrome c 23 ; hemoglobin 19, 39) argue strongly that class II-restricted TCR share a common orientation similar to that observed for class I-restricted TCR.

Although previous experiments examining the ability of class II molecules to positively select CD4+ T cells expressing a structurally related TCR transgene (AR-5,V{alpha}3.5, see Fig. 1Go) indicated that residues 27{alpha} (CDR1) and 51{alpha} (CDR2) probably interacted with MHC residues in an orientation that was preserved for a variety of selecting Ags, this study was unable to define the precise orientation of the TCR on the Ag/MHC complex 40 . Our observation that TCR amino acid residues on either side of 27{alpha} define the ability of BI-specific TCR to recognize MHC ß-chain residues confirms the role of the CDR1 region in recognition of MHC and raises the possibility that selection of V{alpha}3.5-expressing TCR is mediated through interactions predominantly with residues on the MHC ß-chain {alpha}-helix.

Despite the structural constraints imposed by a common orientation on the peptide/MHC complex, T cell recognition was remarkably variable for the different BI-specific TCR. Even TCR that share structural similarity, for example the same {alpha}- or ß-chain sequence, show heterogeneity in their response to the panel of MHC mutants. This heterogeneity appears to arise from two distinct mechanisms. First, interactions between one chain (or CDR) of the TCR and the peptide/MHC complex appear to affect interactions between the other chain (or CDR) of the TCR and the complex. This is most apparent at the center of the peptide binding groove, where we predict that both CDR3 regions are located adjacent to P5. This could result from conformational shifts in the peptide and/or MHC molecule induced by binding to a TCR chain or, alternatively, shared binding of both TCR CDR3 to the same amino acid residue of the peptide, as demonstrated for the TCR/peptide/class I complex, A6/Tax/HLA-A2 5 .

Binding of the TCR CDR1 and/or CDR2 to the peptide/MHC complex also appeared to vary depending upon the amino acid sequence of the CDR3 region. For example, while all V{alpha}3.4-J{alpha}41 hybridomas share the same pattern of responses to carboxyl-terminal MHC mutants, hybridomas expressing V{alpha}3.4 paired with other J{alpha} gene segments do not (Figs. 3Go and 4Go). Similarly, the ability of Vß8.3-Jß1.1 hybridomas to respond to PI presented by W61A is dependent upon the presence of both gene segments (Fig. 6Go). In this connection, crystallography of the A6/Tax/HLA-2 complex shows that CDR1{alpha} and CDR3{alpha} share interactions with P4, while CDR1ß and CDR3ß both interact with P8 5 . Notably, the ability of V{alpha}3.4-J{alpha}41-expressing hybridomas to respond to BI presented by carboxyl-terminal MHC mutants is independent of ß-chain usage. This suggests that although conformational changes are an important source of flexibility in T cell recognition, they appear to predominantly act locally.

The second mechanism that leads to variation in the T cell response to insulin complexed with the different MHC mutants is the finely tuned nature of the threshold for T cell activation. Even minor differences in insulin reactivity can dramatically alter the ability of a hybridoma to respond to subtle changes in the peptide/MHC complex, as reflected in the variable responses of the hybridomas with similar TCR to MHC mutants A78V and H81A or altered peptides PI and SI (Figs. 6Go and 7Go). Although our experiments do not specifically address the mechanism underlying this fine-tuning, we propose that individual interactions between the TCR and Ag/MHC complex probably act in a synergistic fashion to achieve the activation threshold. This is based upon the observation that relatively widely separated interactions, such as the interaction between the TCR and residues A8–10 of the antigenic peptide (where the species variants of insulin differ) located adjacent to W61A at one end of the peptide binding groove, and the TCR and H81A (for the differences between BI and SI reactivity with GBD11, Fig. 7Go) or T77A (acquisition of PI recognition for FBD37, Fig. 5Go) at the other end of the groove can act in concert to achieve the activation threshold. Similar findings were observed by Ehrich et al. in their study of cytochrome c-reactive hybridomas with structurally related TCR. These investigators found that even a single amino acid change in the antigenic peptide had a profound effect on the ability of relatively distant TCR/MHC interactions 41 to support T cell recognition. Indeed, the response of their hybridomas was so variable that they were unable to define a unique orientation for their hybridomas on the Ag/MHC complex.

The findings outlined in this study provide insight into the T cell recognition events that form the basis for thymic selection. TCR recognition of peptide/MHC complexes by developing T cells in the thymus plays a central role in selection of the expressed TCR repertoire. T cells bearing TCR with high affinity for these complexes are deleted, while those with low affinity are exported to the periphery 42, 43, 44, 45 . This process results in a peripheral T cell repertoire that is skewed toward recognition of foreign peptides complexed with self MHC molecules, recently dramatically demonstrated by Ignatowicz et al. 46 . In mice whose T cells were selected on a single peptide/MHC complex (Ab), 65% of CD4+ T cells in the periphery were found to react to Ab complexed with self-peptides, whereas 25% reacted to self-peptides complexed with Ad, the most closely related class II molecule to Ab. Our finding that class II-restricted TCR share a unique orientation provides some insight into the structural basis for this phenomenon. Positive selection of a TCR in the thymus will result in export of a T cell with low affinity for the selecting peptide/MHC complex. The affinity of this TCR for the peptide/MHC complex is the sum of the individual interactions between residues of the TCR and MHC as well as those between the TCR and peptide. In the periphery, because of the structural constraints imposed by a single orientation for the TCR on the complex, some of the low-affinity interactions between MHC and TCR will be preserved despite local conformational changes induced by the antigenic peptide. The net result would be to reduce the binding contribution required from the peptide to achieve the T cell activation threshold. For MHC molecules that differ from the selecting MHC molecule, the absence of these low-affinity preselected interactions will result in an increase in the binding contribution that is required from the antigenic peptide to reach the activation threshold. While this does not preclude T cell recognition events involving nonselecting MHC alleles, it makes them much less likely.


   Acknowledgments
 
We thank Dr. R. Inman for critically reading the manuscript, Novo-Nordisk for generously providing monocomponent beef and pork insulin, and Bhushan Nagar and Dr. Jim Rini for their assistance with three-dimensional visualization of TCR and peptide/MHC crystal structures using the program SETOR (47).


   Footnotes
 
1 This work was supported by a grant from the Medical Research Council of Canada and by a research scholarship from the Arthritis Society of Canada (to J.W.). Back

2 Address correspondence and reprint requests to Dr. Joan Wither, Arthritis Center of Excellence, Toronto Hospital-Western Division, FP1-212, 399 Bathurst St., Toronto, Ontario, Canada M5T 2S8. Back

3 Abbreviations used in this paper: CDR, complementarity-determining region; BI, beef insulin; PI, pork insulin; SI, sheep insulin; EI, equine insulin; CA, conalbumin. Back

Received for publication June 29, 1998. Accepted for publication October 30, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Babbitt, B. P., P. M. Allen, G. Matsueda, E. Haber, E. R. Unanue. 1985. Binding of immunogenic peptides to Ia histocompatibility molecules. Nature 317:359.[Medline]
  2. Kronenberg, M., G. Sui, L. E. Hood, N. Shastri. 1986. The molecular genetics of the T cell antigen receptor and T cell recognition. Annu. Rev. Immunol. 4:529.[Medline]
  3. Davis, M. M., P. J. Bjorkman. 1988. T cell antigen receptor genes and T cell recognition. Nature 334:395.[Medline]
  4. Garcia, K. C., M. Degano, R. L. Stanfield, A. Brunmark, M. R. Jackson, P. A. Peterson, L. Teyton, I. A. Wilson. 1996. An {alpha}ß T cell receptor structure at 2.5Å and its orientation in the TCR-MHC complex. Science 274:209.[Abstract/Free Full Text]
  5. Garboczi, D. N., P. Ghosh, U. Utz, Q. R. Fan, W. E. Biddison, D. C. Wiley. 1996. Structure of the complex between human T-cell receptor, viral peptide and HLA-A2. Nature 384:134.[Medline]
  6. Garcia, K. C., M. Degano, L. R. Pease, M. Huang, P. A. Peterson, L. Teyton, I. A. Wilson. 1998. Structural basis of plasticity in T cell receptor recognition of a self peptide-MHC antigen. Science 279:1166.[Abstract/Free Full Text]
  7. Ding, Y.-H., K. J. Smith, D. N. Garboczi, U. Utz, W. E. Biddison, D. C. Wiley. 1998. Two human T cell receptors bind in a similar diagonal mode to the HLA-A2/Tax peptide complex using different TCR amino acids. Immunity 8:403.[Medline]
  8. Teng, M.-K., A. Smolyar, A. G. D. Tse, J.-H. Liu, J. Liu, R. E. Hussey, S. G. Nathenson, H.-C. Chang, E. L. Reinherz, J.-H. Wang. 1998. Identification of a common docking topology with substantial variation among different TCR-peptide-MHC complexes. Curr. Biol. 8:409.[Medline]
  9. Chothia, C., D. R. Boswell, A. M. Lesk. 1988. The outline structure of the T cell {alpha}ß receptor. EMBO J. 7:3745.[Medline]
  10. Claverie, J. M., A. Prochnicka Chalufour, L. Bougueleret. 1989. Implications of a Fab-like structure for the T cell receptor. Immunol. Today 10:10.[Medline]
  11. Madden, D. R.. 1995. The three-dimensional structure of peptide-MHC complexes. Annu. Rev. Immunol. 13:587.[Medline]
  12. Jardetzky, T. S., J. H. Brown, J. C. Gorga, L. J. Stern, R. G. Urban, J. L. Strominger, D. C. Wiley. 1996. Crystallographic analysis of endogenous peptides associated with HLA-DR1 suggests a common, polyproline II-like conformation for bound peptides. Proc. Natl. Acad. Sci. USA 93:734.[Abstract/Free Full Text]
  13. Garrett, T. P. J., M. A. Saper, P. J. Bjorkman, J. L. Strominger, D. C. Wiley. 1989. Specificity pockets for the side chains of peptide antigens in HLA-Aw68. Nature 342:692.[Medline]
  14. Young, A. C. M., W. Zhang, J. C. Sacchettini, S. G. Nathenson. 1994. The three-dimensional structure of H-2Db at 2.4 Å resolution: implications for antigen-determinant selection. Cell 76:39.[Medline]
  15. Brown, J. H., T. S. Jardetzky, J. C. Gorga, L. J. Stern, R. G. Urban, J. L. Strominger, D. C. Wiley. 1993. Three-dimensional structure of the human class II histocompatibility antigen HLA-DR1. Nature 364:33.[Medline]
  16. Stern, L. J., J. H. Brown, T. S. Jardetzky, J. C. Gorga, R. G. Urban, J. L. Strominger, D. C. Wiley. 1994. Crystal structure of the human class II MHC protein HLA-DR1 complexed with an influenza virus peptide. Nature 368:215.[Medline]
  17. Fremont, D. H., M. Matsumura, E. A. Stura, P. A. Peterson, I. A. Wilson. 1992. Crystal structure of two viral peptides in complex with murine MHC class I H-2Kb. Science 257:919.[Abstract/Free Full Text]
  18. Madden, D. R., D. N. Garboczi, D. C. Wiley. 1993. The antigenic identity of peptide-MHC complexes: a comparison of the conformations of five viral peptides presented by HLA-A2. Cell 75:693.[Medline]
  19. Fremont, D. H., W. A. Hendrickson, P. Marrack, J. Kappler. 1996. Structures of an MHC class II molecule with covalently bound single peptides. Science 272:1001.[Abstract]
  20. Scott, C. A., P. A. Peterson, L. Teyton, I. A. Wilson. 1998. Crystal structures of two I-Ad-peptide complexes reveal that high affinity can be achieved without large anchor residues. Immunity 8:319.[Medline]
  21. Wither, J., J. Pawling, L. Phillips, T. Delovitch, N. Hozumi. 1991. Amino acid residues in the T cell receptor CDR3 determine the antigenic reactivity patterns of insulin-reactive hybridomas. J. Immunol. 146:3513.[Abstract]
  22. Danska, J. S., A. M. Livingstone, V. Paragas, T. Ishihara, C. G. Fathman. 1990. The presumptive CDR3 regions of both T cell receptor {alpha} and ß chains determine T cell specificity for myoglobin peptides. J. Exp. Med. 172:27.[Abstract/Free Full Text]
  23. Jorgensen, J. L., U. Esser, B. Fazekas de St. Groth, P. A. Reay, M. M. Davis. 1992. Mapping T-cell receptor-peptide contacts by variant peptide immunization of single-chain transgenics. Nature 355:224.[Medline]
  24. Sant’Angelo, D. B., G. Waterbury, P. Preston-Hurlburt, S. T. Yoon, R. Medzhitov, S.-C. Hong, C. A. Janeway. 1996. The specificity and orientation of a TCR to its peptide-MHC class II ligand. Immunity 4:367.[Medline]
  25. Sun, R., S. E. Shepherd, S. S. Geier, C. T. Thomson, J. M. Sheil, S. G. Nathenson. 1995. Evidence that the antigen receptors of cytotoxic T lymphocytes interact with a common recognition pattern on the H-2Kb molecule. Immunity 3:573.[Medline]
  26. Turner, S. J., S. C. Jameson, F. R. Carbone. 1997. Functional mapping of the orientation for TCR recognition of an H2-Kb-restricted ovalbumin peptide suggests that the ß-chain subunit can dominate the determination of peptide side chain specificity. J. Immunol. 159:2312.[Abstract/Free Full Text]
  27. Vukusic, B., L. Poplonski, L. Phillips, J. Pawling, T. Delovitch, N. Hozumi, J. Wither. 1995. Both MHC and background gene heterozygosity alter T cell receptor repertoire selection in an antigen-specific response. Mol. Immunol. 32:1355.[Medline]
  28. Glimcher, L. H., D. J. McKean, E. Choi, J. G. Seidman. 1985. Complex regulation of class II gene expression: analysis with class II mutant cell lines. J. Immunol. 135:3542.[Abstract]
  29. Koch, N., G. J. Hammerling, N. Tada, S. Kimura, U. Hammerling. 1982. Cross-blocking studies with monoclonal antibodies against I-A molecule of haplotypes b, d and k. Eur. J. Immunol. 12:909.[Medline]
  30. Jensen, P. E.. 1991. Reduction of disulfide bonds during antigen processing: evidence from a thiol-dependent insulin determinant. J. Exp. Med. 174:1121.[Abstract/Free Full Text]
  31. Williams, D. B., J. Ferguson, J. Gariepy, D. McKay, Y.-T. Teng, S. Iwasaki, N. Hozumi. 1993. Characterization of the insulin A-chain major immunogenic determinant presented by MHC class II I-Ad molecules. J. Immunol. 151:3627.[Abstract]
  32. Brown, J. H., T. Jardetzky, M. A. Saper, B. Samraoui, P. J. Bjorkman, D. C. Wiley. 1988. A hypothetical model of the foreign antigen binding site of class II histocompatibility molecules. Nature 376:845.
  33. Griffith, I. J., N. Nabavi, Z. Ghogawala, C. G. Chase, M. Rodriguez, D. J. McKean, L. H. Glimcher. 1988. Structural mutation affecting intracellular transport and cell surface expression of murine class II molecules. J. Exp. Med. 167:541.[Abstract/Free Full Text]
  34. Becker, D. M., P. Patten, Y.-H. Chien, T. Yakota, Z. Eshhar, M. Giedlin, N. R. J. Gascoigne, C. Goodnow, R. Wolf, K.-I. Arai, et al 1985. Variability and repertoire size of T-cell receptor V{alpha} gene segments. Nature 317:430.[Medline]
  35. Koop, B. F., R. K. Wilson, K. Wang, B. Vernooij, D. Zaller, C. L. Kuo, D. Seto, M. Toda, L. Hood. 1992. Organization, structure, and function of 95kb of DNA spanning the murine T-cell receptor C{alpha}/C{delta} region. Genomics 13:1209.[Medline]
  36. Kabat, E. A., T. T. Wu, H. M. Perry, K. S. Gottesman, C. Foeller. 1991. Sequences of Proteins of Immunological Interest 1506. U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health, Bethesda.
  37. Manning, T. C., C. J. Schlueter, T. C. Brodnicki, E. A. Parke, J. A. Speir, K. C. Garcia, L. Teyton, I. A. Wilson, D. M. Kranz. 1998. Alanine scanning mutagenesis of an {alpha}ß T cell receptor: mapping the energy of antigen recognition. Immunity 8:413.[Medline]
  38. Hong, S.-C., D. B. Sant’Angelo, B. N. Dittel, R. Medzhitov, S. T. Yoon, P. G. Waterbury, C. A. Janeway. 1997. The orientation of a T cell receptor to its MHC class II:peptide ligands. J. Immunol. 159:4395.[Abstract]
  39. Hsu, B. L., B. D. Evavold, P. M. Allen. 1995. Modulation of T cell development by an endogenous altered peptide ligand. J. Exp. Med. 181:805.[Abstract/Free Full Text]
  40. Sim, B.-C., L. Zerva, M. I. Greene, N. R. J. Gascoigne. 1996. Control of MHC restriction by TCR V{alpha} CDR1 and CDR2. Science 273:963.[Abstract]
  41. Ehrich, E. W., B. Devaux, E. P. Rock, J. L. Jorgensen, M. M. Davis, Y.-H. Chien. 1993. T cell receptor interaction with peptide/major histocompatibility complex (MHC) and superantigen/MHC ligands is dominated by antigen. J. Exp. Med. 178:713.[Abstract/Free Full Text]
  42. Sebzda, E., V. A. Wallace, J. Mayer, R. S. M. Yeung, T. W. Mak, P. S. Ohashi. 1994. Positive and negative thymocyte selection induced by different concentrations of a single peptide. Science 263:1615.[Abstract/Free Full Text]
  43. Alam, S. M., P. J. Travers, J. L. Wung, W. Nasholds, S. Redpath, S. C. Jameson, N. R. J. Gascoigne. 1996. T-cell-receptor affinity and thymocyte positive selection. Nature 381:616.[Medline]
  44. Ashton-Rickardt, P. G., A. Bandeira, J. R. Delaney, L. Van Kaer, H.-P. Pircher, R. M. Zinkernagel, S. Tonegawa. 1994. Evidence for a differential avidity model of T cell selection in the thymus. Cell 76:651.[Medline]
  45. Hogquist, K. A., S. C. Jameson, W. R. Heath, J. L. Howard, M. J. Bevan, F. R. Carbone. 1994. T cell receptor antagonist peptides induce positive selection. Cell 76:17.[Medline]
  46. Ignatowicz, L., J. Kappler, P. Marrak. 1996. The repertoire of T cells shaped by a single MHC/peptide ligand. Cell 84:521.[Medline]
  47. Evans, S. V.. 1993. SETOR: hardware-lighted three-dimensional solid model representations of macromolecules. J. Mol. Graphics 11:134.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wither, J. E.
Right arrow Articles by Vukusic, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wither, J. E.
Right arrow Articles by Vukusic, B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS