| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Microbiology and Immunology and
Cell Biology, Albert Einstein College of Medicine, Bronx, NY 10461
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain transgenic mice with the VSV peptide and raised
a panel of anti-VSV CTL clones with identical TCR -chains.
Consistent with our previous analysis of uncloned populations of
primary CTLs, the anti-VSV CTL clones were all Vß13+
and expressed TCR ß-chains with highly homologous
complementarity-determining region 3 (CDR3) loops. Although the clones
expressed similar TCRs, they differed in their ability to cross-react
with VSV peptide variants singly substituted at TCR contact positions 4
and 6. These findings allowed us to identify short stretches of amino
acids in the C-terminal region of the CDR3ß loop that, when altered,
modify the cross-reaction capability of the TCR to position 4 and
position 6 variant peptides. To further probe the structural correlates
of biologic cross-reactivity, we used cross-reactive CTL clones and
cell lines expressing point mutations in H-2Kb to
investigate the effect of single amino acid changes in the peptide on
the pattern of recognition of the TCR for the peptide/MHC complex.
Single conservative substitutions in the peptide were sufficient to
alter the recognition contacts between a cross-reactive TCR and the MHC
molecule, supporting the idea that the TCR can make overall structural
adjustments in MHC contacts to accommodate single amino acid changes in
the peptide. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß TCRs respond to a vast array of foreign Ags that are
presented as peptides bound to MHC class I or class II molecules
(1, 2, 3). MHC class I molecules are expressed on most cell types, and
they present primarily endogenous cytosolic peptides to
CD8+ CTLs. In contrast, MHC class II molecules are
expressed mainly on APCs, such as dendritic cells, macrophages, and B
cells, and they present peptides derived from exogenous Ags to
CD4+ Th cells.
Functional TCR - and ß-chain genes determining the TCRs that
recognize the peptide/MHC targets arise from somatic rearrangement of
germline V (variable), D (diversity; ß-chain only), and J (joining)
gene segments. Each chain of the TCR ß heterodimer has three
hypervariable regions that are analogous to the Ag-binding
complementarity-determining regions
(CDRs)5 of Igs and are
similarly referred to as CDR1, CDR2, and CDR3. The CDR1 and CDR2 loops
of each TCR chain are encoded by the germline V gene segments, while
CDR3 is encoded at the V-J junction in the -chain and at the V-D-J
junction in the ß-chain (4, 5). The specificity for recognizing
Ag comes from the combinatorial mechanism of somatic gene
rearrangement, which is partly responsible for the diversity found in
the T cell repertoire, and from diversity arising because the joining
events that form the junctions between the gene segments are imprecise,
and nontemplate-encoded (N) nucleotides can be added at the junctions.
As the CDR3 loops are encoded at these junctions, they are the most
diverse regions of TCR - and ß-chains (4) and play a major role in
recognizing Ag (6, 7, 8, 9).
The solution of the crystal structures of three different class
I-restricted TCRs complexed with their respective peptide/MHC ligands
has considerably enriched our understanding of the features of
TCR-ligand interactions (6, 7, 8, 9). Although there are significant
differences among the solved structures, they have revealed certain
general features of the interaction between a TCR and its ligand. In
all the structures, the TCR is oriented diagonally over the peptide/MHC
complex, with the -chain over the N-terminal region of the peptide
and the ß-chain over the C-terminal region of the peptide. CDR2
appears over the 2 helix of the MHC class I molecule, and CDR2ß
appears over the 1 helix. CDR1, CDR3, and CDR3ß all contact
the peptide, with the CDR3 loops positioned near the central residue of
the peptide. However, the solved structures differ in the extent of
interaction that occurs between the peptide and the CDR1 and CDR3 loops
of the ß-chain. Also, in one of the crystal structures, all six of
the CDR loops form contacts with MHC residues (8), but in the others,
the CDR1 and CDR2 loops of the ß-chain make few or no contacts with
the MHC molecule (7, 9). The observed variation in the details of the
interaction between TCRs and their ligands underscores the need for
more studies, both biologic and three-dimensional, to further define
the features of this crucial immunologic interaction.
A number of TCRs, like the recently crystallized 2C TCR, can recognize two or more different MHC molecules. This finding demonstrates the property of cross-reactivity, the basis for the phenomenon of alloreaction (10, 11, 12). Cross-reactivity to more similar ligands occurs when a single TCR recognizes multiple peptides with limited sequence homology but presented in the context of a single MHC molecule (13, 14, 15, 16, 17). This cross-reaction capability helps to explain how T cells selected in the thymus to recognize self peptides can nonetheless later function peripherally in the recognition of foreign peptides derived from pathogens. Cross-reactivity also may be involved in the induction or worsening of an autoimmune disease by a viral infection (16, 18). The least extreme case of cross-reactivity occurs when a single TCR can recognize not only its cognate peptide/MHC, but also certain singly substituted peptide variants. Such peptide variants sometimes constitute epitopes that are functionally equivalent to the original peptide, or they may act as antagonists, partial agonists, or superagonists (19, 20, 21). The TCR sequences that enable one TCR but not another to recognize singly substituted peptide variants have not been well characterized. It is also not known whether conservative changes in single peptide residues can alter the recognition pattern of a cross-reactive TCR on its peptide/MHC ligand.
In this study we address the issue of the structural correlates of
biologic cross-reactivity in its simplest form, the cross-recognition
of singly substituted peptide variants. To do this, we use a
well-characterized class I MHC system in which the VSV octapeptide (the
immunodominant epitope of the vesicular stomatitis virus nucleocapsid
protein with the sequence RGYVYQGL) is recognized by CTLs in the
context of H-2Kb (22, 23). Functional and crystal
structural studies of VSV/H-2Kb have revealed that the
amino acid residues of VSV at positions 3 (Tyr), 5 (Tyr), and 8 (Leu)
are the anchor residues, while positions 1 (Arg), 4 (Tyr), and 6 (Gln)
are solvent exposed and serve as TCR contact residues (24, 25, 26). To
precisely analyze TCR CDR3ß structures and their interaction with the
VSV peptide and Kb, we recently generated TCR -chain
transgenic mice in a TCR -deficient background using the TCR
-chain from a VSV peptide-specific CTL clone as the transgene. Using
these mice, we obtained in vivo evidence for specific
interaction between the amino acid residue at position ß98 of the TCR
CDR3ß loop and position 6 of the VSV peptide or its variants (27).
Immunization of our transgenic mice with the VSV peptide or its
position 6 variant peptides elicited primary CTLs with highly
homologous TCR ß-chains, as determined by RT-PCR analysis (27). The
VSV peptide elicited Vß13+ CTLs with a conserved
Gly-Val/Thr motif at ß97-ß98 of the CDR3ß loop, while
immunization with peptide variants containing Asp or Glu at position 6
yielded Vß7+ CTLs with a conserved Arg at ß98. Thus,
RT-PCR analysis of uncloned populations of primary CTLs was sufficient
to enable us to identify conserved features of TCR ß-chains important
for the recognition of a specific peptide. However, further work
investigating the structural basis of cross-reactivity required the
availability of T cell clones whose fine peptide specificities and MHC
recognition patterns could be analyzed in detail. In this study, we
immunized the TCR -chain transgenic mice with the VSV peptide or its
variants and then raised panels of CTL clones. Analysis of each
clone’s TCR CDR3ß sequence and its specificity to peptides singly
substituted at the TCR contacting residues allowed us to identify TCR
ß-chain residues that can alter the cross-reaction capability to
position 4 and position 6 variant peptides. Further, comparison of the
recognition patterns of a cross-reactive TCR on different
peptide/H-2Kb ligands revealed that subtle single amino
acid changes in the peptide can dramatically alter the recognition
contacts between a TCR and its class I MHC ligand.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The derivation of TCR-/-TgV2+
transgenic mice was described in detail previously (27). These mice are
transgenic for the TCR -chain of the
V2+Vß13+ anti-VSV CTL clone N30.7
(23). Because they also carry a targeted disruption of both alleles of
the endogenous TCR C locus, they cannot express their endogenous TCR
-chain genes. C57BL/6 (B6) mice were purchased from The Jackson
Laboratory (Bar Harbor, ME). All mice were maintained in the Animal
Institute of the Albert Einstein College of Medicine.
Cell lines
R8 is a heterozygous (H-2bxd) pre-B lymphoblastoid cell line transformed by Abelson leukemia virus. R8 cell lines containing point mutations of H-2Kb have been previously described (23, 28, 29, 30, 31). The mutant R8 lines used in this study and the location and identity of their H-2Kb mutations are as follows: R8.8 (G56E), R8.9 (E58K), R8.62 (R62Q), R8.110.43 (R75Q), R8.18 (R79K), R8.24 (T80I), R8.313 (L82P), R8.208 (L82F), R8.125 (G90D), R8.127 (M138K), R8.14 (L141R), R8.331 (A150P), R8.110.2 (A158T), R8.34 (G162D), R8.347 (E166K), R8.10 (W167R), and R8.353 (N174K). Wild-type R8 cells and all H-2Kb mutants were maintained in DMEM supplemented with 5% heat-inactivated FCS. The H-2Kb mutants expressed comparable levels of MHC molecules relative to the wild-type R8 cells as revealed by flow cytometric analysis (32).
Peptide synthesis and purification
VSV peptide (RGYVYQGL) and all singly substituted variant peptides were synthesized by standard solid phase methods using F-moc chemistry in a peptide synthesizer (model 433A, Applied Biosystems, Foster City, CA) at the Peptide Synthesis Facility of the Albert Einstein College of Medicine. The peptides were cleaved from the resin, and the side chain protecting groups were removed using 95% trifluoroacetic acid. They were purified to >98% homogeneity by reverse phase HPLC on a Vydac C18 semipreparative column (218TP510, Vydac, Hesperia, CA). The identity of the purified peptides was confirmed using a tandem quadrupole mass spectrometer (TSQ700, Finnigan MAT, San Jose, CA).
Establishment and maintenance of CTL clones
TCR-/-TgV2+ transgenic mice were
immunized in their hind footpads with 15 µg of peptide emulsified in
CFA. The mice were boosted once with 15 µg of peptide emulsified in
IFA 1 wk after the primary immunization. One week after the booster,
spleen tissues were removed, and spleen cells (5 x
107) were cultured in 10 ml of culture medium with 1
x 10-6 M peptide in tissue culture flasks (Falcon 3082,
Becton Dickinson, Franklin Lakes, NJ) for 7 days at 37°C under 9%
CO2 in air. The culture medium used was Iscove’s modified
Dulbecco’s medium supplemented with 10% heat-inactivated FCS
(HyClone, Logan, UT), 2 mM glutamine, 50 U/ml penicillin, 50 µg/ml
streptomycin, and 50 µM 2-ME. The harvested cells (1 x
105) were passaged and maintained by weekly restimulation
with 2 x 10-7 M peptide and mitomycin C (Sigma, St.
Louis, MO)-treated B6 spleen cells (5 x 106) in
complete medium supplemented with 50 U/ml human recombinant IL-2 (Life
Technologies, Gaithersburg, MD) in 24-well culture plates (Falcon 3047,
Becton Dickinson). CTL clones were derived from CTL lines by limiting
dilution in the presence of 2 x 10-7 M peptide and
mitomycin C-treated B6 spleen cells (1 x 106) in
96-well culture plates (3799, Costar, Cambridge, MA). Clones were
tested for cytotoxicity and maintained by weekly restimulation as
described for CTL lines.
Cell-mediated cytotoxicity assay
R8 and its H-2Kb point mutants were labeled with 1.85 MBq of Na251CrO4 (Amersham, Arlington Heights, IL) for 1 h at 37°C under 9% CO2 in air. They were washed and used as target cells. In sensitization assays, 10 µl of peptide solution was added to 5 x 103 51Cr-labeled target cells (100 µl) and incubated for 1 h at room temperature before adding effector cells (100 µl). Final peptide concentrations are given in the figure legends. After incubating target and effector cells for 4 h at 37°C under 9% CO2 in air, the supernatants (100 µl) were removed, and their radioactivities were measured with a gamma counter. The percent specific lysis was calculated as 100 x (experimental release - spontaneous release)/(maximum release - spontaneous release), where experimental release is the radioactivity in the supernatant of target cells mixed with effector cells, spontaneous release is that in the supernatant of target cells incubated alone, and maximum release is that in the supernatant after lysis of target cells with 2% Triton X-100.
RT-PCR and TCR ß-chain sequence determination
mRNA was purified from approximately 1 x 106 Ficoll-purified CTLs using the QuickPrep Micro mRNA Purification Kit (Pharmacia, Piscataway, NJ). The mRNA was reverse transcribed into single strand cDNA using Moloney murine leukemia virus reverse transcriptase and oligo(dT)15 as primer. Vß7 or Vß13 TCR cDNAs were amplified by PCR using a Vß7-specific primer (5'-TACAGGGTCTCACGGAAGAAGC-3') or a Vß13-specific primer (5'-AGGCCTAAAGGAGCTAACTCCAC-3'), paired in each case with a Cß primer (5'-CACTGATGTTCTGTGTGACAG-3'). PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, Valencia, CA), and the CDR3 regions were sequenced from both directions at the DNA Sequencing Facility of the Albert Einstein College of Medicine.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We previously generated TCR -chain transgenic mice in a TCR
-deficient background using as the transgene the -chain from the
VSV peptide-specific CTL clone N30.7. We showed that immunization of
these mice with the VSV peptide enabled us to generate CTL lines that
could recognize and lyse target cells pulsed with the VSV peptide (27).
From such lines, we have now raised four CTL clones, all of which are
CD8+ T cells as revealed by flow cytometry (data not
shown). All the CTL clones recognize VSV peptide-pulsed R8 target cells
that express H-2Kb (Fig. 1
).
|
). All the CTL clones generated from VSV
immunization used the Vß13 gene segment and had Gly and Val at
positions ß97 and ß98 of the CDR3 loop, respectively. These results
are consistent with our previous study in which we showed that the
Vß13 gene segment is used by >80% of primary CTLs generated by VSV
immunization of our transgenic mice, and that the CDR3ß Gly-Val motif
is highly conserved in such CTLs (27). Two of the four clones had TCR
ß-chain sequences identical with those identified by RT-PCR analysis
of primary CTLs reported in this previous study. Three of the four
clones had a CDR3ß loop of nine amino acids, the preferred length for
Vß13+ VSV-specific CTLs from the transgenic mice,
although one (Tg.W2B) had 10 amino acids. Note that the TCR ß-chain
of Tg.W3A was almost identical with that of the N30.7 CTL clone, which
was the source of the V2 transgene, except for the two amino acids
at the N-terminal end of the CDR3ß loop.
|
In our previous structural (25) and functional studies (24),
positions 1, 4, and 6 of the VSV peptide were shown to be exposed for
potential TCR interaction. Gln at position 6 is an important TCR
contact residue that interacts with the TCR ß-chain (27). We
investigated the ability of our CTL clones to cross-react with position
6 variant peptides (Fig. 1
). Three of the four clones (Tg.W2B, Tg.W3A,
and Tg.W7A) were specific for the VSV peptide and did not recognize any
position 6 variant peptides, including those containing negatively
charged amino acid substitutions (D6, Gln to Asp; E6, Gln to Glu),
positively charged amino acid substitutions (K6, Gln to Lys; R6, Gln to
Arg), or a hydrophobic amino acid substitution (I6, Gln to Ile). Only
clone Tg.W4A was cross-reactive, being able to recognize the D6 and E6
peptides bound to H-2Kb, although its reactivity to D6 was
weak. Clones Tg.W4A, Tg.W7A, and Tg.W2B all expressed the transgenic
TCR -chain and had TCR ß-chains that were identical from the amino
terminus up to and including ß98 (Fig. 2). However, only Tg.W4A
showed cross-reactivity to the variant peptides D6 and E6. These
results indicate that the amino acid residues C-terminal to position 98
of the TCR ß-chain are important in determining the capability for
cross-reaction to position 6 variant peptides.
To evaluate the ability of the four anti-VSV clones to cross-react
with peptides substituted at the two other potential TCR contact sites,
peptide variants with substitutions at position 4 or position 1 were
analyzed for the ability to sensitize target cells to lysis by the CTL
clones. While most substitutions at position 4 were not well tolerated,
peptides with conservative substitutions were recognized by some of the
clones (Fig. 3). A4 (Val to Ala) was
permissive for recognition by Tg.W3A, Tg.W4A and Tg.W7A, while I4,
containing a conservative methylene group addition (Val to Ile), was
permissive for recognition by Tg.W2B and Tg.W3A. Tg.W7A and Tg.W2B have
nearly identical TCR ß-chains (Fig. 2), except that Tg.W7A has
Ser-Tyr at ß99 and ß100, while Tg.W2B has Gly-Gly-Arg at ß99 to
ß101. Despite this extensive sequence similarity, they reacted
differently to position 4 variant peptides. These results suggest that
the CDR3ß amino acid residues C-terminal to ß98 are important in
determining the capability for cross-reaction to position 4 variant
peptides.
|
). Overall, no striking differences were seen among the
four clones in the ability to recognize position 1 variant peptides,
although Tg.W7A recognized the M1/Y1 and N1/H1 peptide mixtures very
weakly. Clones with very similar TCR ß-chains show different recognition patterns on VSV/H-2Kb
A number of R8 cell lines containing point mutations of
H-2Kb were previously selected using mAbs (29, 31) or CTLs
(28, 30), and nearly all carry nonconservative substitutions in the
1 or 2 helix of the Kb molecule. For each of the
anti-VSV clones, these R8 cell lines were used to identify contacts
made between the TCR and its ligand during the process of recognition.
The R8 cell lines were assessed for recognition by each CTL clone, and
loss of recognition of a cell line expressing Kb with a
specific mutation was assumed to reflect a TCR-Kb contact
point at the mutated residue. Responses of the anti-VSV CTL clones
to the H-2Kb mutants are shown in Fig. 4. The recognition pattern of N30.7 is
also shown (23). Most of the clones were only affected by mutations
located on the 2 helix of the Kb molecule, and common
contact points for all the TCRs were identified at residues 166,
167, and 174. Tg.W3A, whose entire TCR ß-chain differs from that of
N30.7 by only two amino acids in CDR3ß (positions 95 and 96), showed
a similar, but not identical, recognition pattern on
VSV/H-2Kb as did N30.7. Likewise, Tg.W2B and Tg.W7A, which
also have highly homologous TCR ß-chains (Fig. 2), nonetheless showed
different TCR recognition patterns. These results show that CTL clones
with highly homologous TCR ß-chains can have different recognition
patterns on the same peptide/MHC class I complex. Conversely, clones
with more dissimilar TCR ß-chains can show identical recognition
patterns. For example, Tg.W3A and Tg.W4A, which use different Jß
segments and share only two amino acids in CDR3ß, showed identical
recognition patterns on the H-2Kb mutants.
|
We measured the responses of the cross-reactive Tg.W4A clone to
the VSV peptide and several variant peptides presented by each of the
H-2Kb mutants to assess the TCR-MHC interactions that occur
during cross-recognition events. The recognition patterns diagrammed in
Fig. 5A show that certain
mutations on the 1 helix of H-2Kb differentially
affected the recognition of these peptides by Tg.W4A, demonstrating
that these MHC residues are differentially contacted by the TCR
depending on the peptide present in the peptide binding groove.
(Quantitative relative lysis data for relevant 1 mutants are
provided in Fig. 5B.) The differences in the recognition
patterns of VSV/H-2Kb and E6/H-2Kb shown in
Fig. 5A are particularly noteworthy, as we have shown that
these ligands are recognized equally well by Tg.W4A in cytotoxicity
assays over a broad range of peptide concentrations (Fig. 1).
|
In an attempt to raise a second panel of CTL clones expressing
identical TCR -chains and highly homologous ß-chains, we immunized
the TCR-/-TgV2+ transgenic mice
individually with D6 or E6 peptide and derived CTL clones capable of
recognizing the immunogen bound to H-2Kb (Fig. 6). The anti-D6 and anti-E6
clones expressed the Vß7 gene segment, and their TCR ß-chains had
Ser-Leu at positions ß95-ß96 in the N-terminal portion of the
CDR3ß loop and a conserved Arg at position ß98 (Fig. 7). We previously showed the importance
of the Arg at ß98 in the interaction with the negatively charged
residue at position 6 of the D6 and E6 peptides (27). Two of the three
CTL clones had ß-chain sequences identical with those identified in
primary CTLs in this earlier study.
|
|
The anti-D6 and anti-E6 CTL clones were all cross-reactive
to some position 6 peptide variants, but their fine specificities
differed (Fig. 6). Tg.D6.1F and Tg.D6.2A showed different specificities
to position 6 variant peptides even though they had identical TCR
ß-chain sequences up to and including ß99 (Fig. 7). Supporting the
conclusion drawn from analysis of the anti-VSV peptide clones,
these results again indicate that the amino acid residues C-terminal to
ß99 are important in determining the capability for cross-recognition
of position 6 peptide variants.
The cross-reactive anti-D6 and anti-E6 clones demonstrate that single amino acid changes in the peptide can alter the recognition pattern of a cross-reactive TCR on its peptide/MHC ligand
For each of the anti-D6 and anti-E6 clones, comparison of
the recognition patterns of peptide/MHC using the immunogen peptides
and the cross-reactive peptides revealed marked differences depending
on the peptide being recognized (Fig. 8).
For example, the Tg.D6.1F TCR contacted position 62 on the 1 helix
of Kb when recognizing the E6 peptide but not when
recognizing the D6 peptide. Conversely, position 80 was contacted
during D6 recognition by this TCR, but not during E6 recognition.
Confirming the results obtained for the anti-VSV clone Tg.W4A (Fig. 5A), the recognition patterns diagrammed in Fig. 8 indicate
that a single TCR can interact differently with the class I MHC
molecule depending on the peptide being recognized, and that its
recognition pattern can be affected by even subtle changes in a single
residue of the peptide.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We previously demonstrated that VSV peptide-specific T cell clones
from wild-type mice with quite diverse TCRs showed different
recognition patterns on VSV/H-2Kb (23). In contrast to CTL
clones generated from immunization of wild-type mice, the
anti-VSV clones that we raised in this study expressed identical
TCR -chains and highly homologous ß-chains. Nonetheless, we again
found that each anti-VSV clone showed an individual recognition
pattern on its H-2Kb ligand (Fig. 4). Further, our data
show that small sequence alterations in either the N- or C-terminal
portions of the CDR3ß loop are sufficient to alter the recognition
pattern of a TCR on its ligand. Additional examples of individual
recognition patterns were provided by our anti-D6 clones (Fig. 8).
The observation that a small change in the sequence of the TCR
ß-chain can alter the recognition pattern of a TCR on its ligand has
also been made in the class II MHC system (33).
Cross-reactivity to position 4 and position 6 variant peptides can be mapped to short segments of CDR3ß
In the VSV/H-2Kb system, we have previously shown that
a number of anti-VSV CTL clones derived from wild-type B6 mice can
cross-react with certain VSV peptide variants substituted at TCR
contact position 1, 4, or 6 (23, 24). Different clones show different
sensitivities to such substitutions in terms of both the positions that
can be altered and the particular substitutions that are tolerated.
However, the lack of homology between the TCRs of these clones
precludes identification of the TCR sequences that are responsible for
this differential cross-reaction capability. The presence of two
productively rearranged TCR -chains in some of the clones further
compounds this problem.
In contrast to CTL clones generated from wild-type mice, the
anti-VSV clones described in this study had highly homologous TCR
ß-chains that were paired exclusively with the transgenic TCR
-chain. Two such clones (Tg.W7A and Tg.W2B) showed different
patterns of cross-reaction to position 4 variant peptides (Fig. 3),
with Tg.W7A reacting only to A4 and Tg.W2B reacting only to I4. These
two clones have identical TCR ß-chains, except that Tg.W7A has
Ser-Tyr at ß99-ß100 near the middle of the CDR3ß loop, while
Tg.W2B has an extra amino acid in this region and the sequence
Gly-Gly-Arg (Fig. 2). These findings allow us to conclude that changes
in this short stretch of amino acids can alter the cross-recognition of
a TCR to position 4 variant peptides, suggesting the possibility of a
direct interaction between this region of CDR3ß and position 4 of the
peptide. Support for such an interaction is provided by the crystal
structures of the A6 and B7 TCRs complexed with the Tax 9-mer peptide
and HLA-A2, in which the central residue of the peptide is found in a
pocket between the CDR3 and CDR3ß loops, and both loops form
contacts with the central residue of the peptide (7, 9).
Our results show that cross-reactivity to position 6 variant peptides
can also be mapped to a short segment of CDR3ß. Tg.W4A and Tg.W7A are
CTL clones that exhibit different abilities to cross-react with
position 6 variants of the VSV peptide (Fig. 1), yet both of the clones
are Vß13+ and have TCRs with CDR3ß loops that are
identical from positions ß95–ß98 (Fig. 2). These findings indicate
that the amino acid residues C-terminal to ß98 in the CDR3ß loop
are important in determining cross-reaction to peptides altered at
position 6. Although the two TCRs use different Jß gene segments, it
is unlikely that J-encoded amino acids C-terminal to the CDR3ß loop
also contribute to this differential cross-reaction, as these amino
acids are framework residues. Further, seven of these 10 J-encoded
amino acids are identical in the two TCRs, and another is a
conservative Ile to Leu change (34). Results obtained with our
anti-D6 clones also indicate the importance of the amino acid
residues C-terminal to position ß99 in the CDR3ß loop in
influencing cross-recognition to position 6 variant peptides (Figs. 6 and 7). These results are consistent with our previous study in which
we demonstrated an interaction between the CDR3ß loop and position 6
of the peptide (27).
The observation made in our class I MHC system that changes in a
limited region of a CDR3 loop can alter the cross-reaction ability of a
TCR was also made for the class II system by Hsu et al. (35). They
studied the cross-recognition capabilities of
Hb64–76/I-Ek-reactive T cell
hybridomas from TCR ß-chain transgenic mice and showed that a change
of only two amino acids in the N-terminus of the CDR3 loop of the
-chain altered the fine specificity of the hybridomas for peptide
variants at the central positions of the peptide.
Single conservative amino acid substitutions in the peptide are sufficient to alter the contacts between a cross-reactive TCR and the class I MHC molecule
The anti-VSV CTL clone Tg.W4A cross-reacts with certain singly
substituted VSV peptide variants, including K1, A4, D6, and E6 (Figs. 1 and 3). Its cross-reactive TCR exhibits different recognition patterns
on a class I MHC molecule depending on the peptide that is present in
the peptide binding groove of the class I molecule (Fig. 5A). We have found that recognition contacts can be altered
by single substitutions at any of the three TCR contact residues of the
VSV peptide, and that even relatively conservative amino acid changes
(e.g., Arg to Lys in K1, Val to Ala in A4, Gln to Glu in E6, or Gln to
Asp in D6) can lead to dramatically altered recognition patterns (Fig. 5A). Results obtained with our three cross-reactive
anti-D6 and anti-E6 CTL clones provide further evidence that
single conservative amino acid substitutions in the peptide are
sufficient to alter the recognition pattern of a cross-reactive TCR on
its peptide/MHC ligand (Fig. 8).
The idea that a single TCR can contact the same MHC in different ways
was first proposed by Ehrich et al. in the class II MHC system (33).
They described two CD4+ T cell hybridomas that recognized
both a murine cytochrome C peptide (amino acids 88–103; 99K) and its
99E variant presented in the context of the class II MHC molecule
I-Ek. For both hybridomas, the interaction with a panel of
I-Ek mutants was found to be different depending on which
peptide was being recognized. However, it was difficult to evaluate the
significance of these findings, because the two hybridomas were
approximately 100 times more sensitive to 99K than to 99E (containing
Lys to Glu, which is a nonconservative change in terms of both size and
charge). We were able to overcome this problem by identifying peptide
variants that were well recognized by our cross-reactive CTL clones.
This allowed us to show that even ligands that are recognized equally
well by a CTL clone can be contacted differently by its TCR. For
example, VSV and E6 are recognized to an equal extent by Tg.W4A (Fig. 1), yet the recognition patterns of Tg.W4A for VSV/Kb and
E6/Kb are quite different (Fig. 5A).
In our experiments, for a given cross-reactive clone, peptide-dependent
alterations in the recognition patterns were located primarily on the
1 helix (Figs. 5A and 8). Only for Tg.D6.2A did a
mutation on the 2 helix (at residue 166) abolish the recognition of
one peptide while not affecting the recognition of another. These
results are in contrast to those reported in the class II MHC system by
Ehrich et al. (33). In their report MHC mutations that abolished the
recognition of a given cross-reactive hybridoma for either 99E or 99K
(but not both) were all found on the ß1 helix, which is analogous to
the 2 helix of the class I molecule. This difference suggests the
intriguing possibility that a given TCR may adapt to slightly altered
peptide ligands in different ways depending on whether the peptide is
bound to a class I or a class II molecule.
There are reasons to believe that class I- and class II-restricted TCRs might bind their ligands in different ways. First, class I- and class II-restricted T cells use different MHC-binding coreceptors (CD8 or CD4, respectively). Second, although class I and class II MHC molecules have remarkably similar structures, there are some differences that are likely to be relevant to T cell recognition. For example, the class II molecule has a peptide-binding site that is open at both ends (36, 37). As a result, the N- and C-termini of the peptide can protrude out of the groove, allowing class II molecules to bind peptides having a variety of lengths (10–34 amino acids) (38, 39, 40). In contrast, the N- and C-termini of peptides bound to class I molecules form hydrogen bonds with conserved MHC residues and are buried at the ends of the binding groove (25, 26, 41, 42). For this reason, class I MHC molecules bind peptides of more defined length (8–10 amino acids) (22, 43, 44) than do class II molecules. Also relevant to the issue of T cell recognition, the total exposed peptide surface is larger for class II-bound peptides than for class I-bound peptides (45). Further, for peptides bound to class I, the exposed peptide surface is largely at the middle of the peptide, whereas it is distributed more uniformly across class II-bound peptides (45). In addition, Carson et al. (46) recently presented striking evidence that peptide-flanking residues of class II-bound peptides that lie outside the minimal epitope that is required to bind to the class II molecule can be directly recognized by certain TCRs and can even act as dominant TCR contact residues. Taken together, this information lends support to the idea that cross-reactive TCRs might adapt to variant peptide/MHC complexes in different ways depending on which class of MHC molecule is in the complex.
We have shown that single amino acid substitutions of TCR contact
residues in an antigenic peptide can alter the recognition pattern of a
cross-reactive TCR on its peptide/class I MHC ligand. These altered
recognition patterns could potentially be due to peptide-induced
changes in the conformation of the MHC molecule. Previous functional
studies using peptide variants containing single substitutions of
TCR-inaccessible residues suggested the occurrence of peptide-specific
alterations in MHC conformation (47, 48). Structural evidence for such
a phenomenon was provided by Fremont and colleagues (49), who compared
the crystal structures of H-2Kb complexed with three
unrelated peptides, OVA-8 (SIINFEKL), VSV-8 (RGYVYQGL), and Sendai
virus-9 (FAPGNYPAL). They found that four MHC side chains took on
peptide-specific conformations (Lys66, Glu152,
Arg155, and Trp167). Similarly, when Madden et
al. (42) crystallized the human class I MHC molecule HLA-A2 with four
different, unrelated 9-mer peptides or a 10-mer, they found three MHC
residues (Arg97, Tyr116, and
Trp167) that could exhibit peptide-specific conformations.
For both H-2Kb and HLA-A2, structural comparisons in the
presence of different peptides revealed a slight main chain shift at
the N-terminal portion of the 2 helix (amino acids 144–151).
Therefore, when a peptide is changed to an unrelated one, minor
conformational changes can occur in the MHC molecule that could
potentially be recognized by a TCR.
In contrast to the unrelated peptides studied by Fremont et al. (49)
and Madden et al. (42), the peptide variants that we studied contained
only single amino acid substitutions of TCR contact residues. Reid et
al. (50) recently compared the crystal structures of HLA-B8 complexed
with an HIV-1 P17 8-mer (GGKKKYKL) or two singly substituted variants
(7Q, GGKKKYQL; 7R, GGKKKYRL) altered at position 7, a TCR contact
point. When the three different structures were compared, the only
changes observed were alterations in the mobilities of the position 7
side chain and two local MHC side chains on the 1 helix
(Glu76 and Asn80). This result then suggests to
us that the VSV peptide and its singly substituted variants are
unlikely to form peptide/MHC complexes with dramatically different
conformations, although crystal structures of H-2Kb with
each of the variant peptides are needed to confirm this suggestion.
Instead, depending on the particular substitution that is made in the
peptide, a change in a single amino acid would alter the charge, size,
and/or shape of the side chain available for contact with the TCR, and
this change might also alter the orientations of neighboring side
chains on the peptide or the MHC molecule (50). The altered recognition
patterns that we observed for a given cross-reactive TCR on different
peptide/H-2Kb complexes are likely to reflect an adaptive
response by the TCR to such subtle local changes at or near the altered
peptide residue. The ability of the CDR loops of a TCR to take on
alternate conformations was recently demonstrated by Garcia et al. (8)
when they compared the crystal structures of the 2C TCR alone or
complexed with its ligand. It is therefore reasonable to hypothesize
that a cross-reactive TCR adjusts the conformation of the CDR loops to
accommodate several different, but closely related, ligands. Our
finding that single conservative substitutions in the peptide are
sufficient to alter the recognition contacts between a cross-reactive
TCR and the class I MHC molecule is consistent with this
hypothesis.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Parasitology and Immunology, Okayama University Medical School, 2-5-1 Shikata-cho, Okayama 700, Japan.
3 Current address: Corixa Corp., 1124 Columbia St., Suite 225, Seattle, WA 98104.
4 Address correspondence and reprint requests to Dr. Stanley G. Nathenson, Department of Microbiology and Immunology, Albert Einstein College of Medicine, 1300 Morris Park Ave., Bronx, NY 10461. E-mail address:
5 Abbreviations used in this paper: CDR, complementarity-determining region; VSV, vesicular stomatitis virus; B6, C57BL/6.
Received for publication April 23, 1998. Accepted for publication July 10, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß T cell receptor structure at 2. 5 Å and its orientation in the TCR-MHC complex. Science 274:209.-helices of the class I MHC molecule are simultaneously recognized by the T cell receptor. Cell 54:47.[Medline]
1 leukemia recognized by cytotoxic T lymphocytes and its relation to the akt oncogene. J. Exp. Med. 180:1599.This article has been cited by other articles:
|
|
 |
|
  J. Lustgarten, A. L. Dominguez, and C. Pinilla Identification of Cross-Reactive Peptides Using Combinatorial Libraries Circumvents Tolerance against Her-2/neu-Immunodominant Epitope J. Immunol., February 1, 2006; 176(3): 1796 - 1805. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. W. Mullins, T. N. J. Bullock, T. A. Colella, V. V. Robila, and V. H. Engelhard Immune Responses to the HLA-A*0201-Restricted Epitopes of Tyrosinase and Glycoprotein 100 Enable Control of Melanoma Outgrowth in HLA-A*0201-Transgenic Mice J. Immunol., November 1, 2001; 167(9): 4853 - 4860. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Honda, W. Zhang, A. M. Kalergis, T. P. DiLorenzo, F. Wang, and S. G. Nathenson Hapten Addition to an MHC Class I-Binding Peptide Causes Substantial Adjustments of the TCR Structure of the Responding CD8+ T Cells J. Immunol., October 15, 2001; 167(8): 4276 - 4285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Zhang, S. Honda, F. Wang, T. P. DiLorenzo, A. M. Kalergis, D. A. Ostrov, and S. G. Nathenson Immunobiological Analysis of TCR Single-Chain Transgenic Mice Reveals New Possibilities for Interaction between CDR3{alpha} and an Antigenic Peptide Bound to MHC Class I J. Immunol., October 15, 2001; 167(8): 4396 - 4404. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. T. Thomson, A. M. Kalergis, J. C. Sacchettini, and S. G. Nathenson A Structural Difference Limited to One Residue of the Antigenic Peptide Can Profoundly Alter the Biological Outcome of the TCR-Peptide/MHC Class I Interaction J. Immunol., March 15, 2001; 166(6): 3994 - 3997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Rosa, R. Krishnan, S. Markel, J. P. Schneck, R. Houghten, C. Pinilla, and D. J. Diamond Enhanced immune activity of cytotoxic T-lymphocyte epitope analogs derived from positional scanning synthetic combinatorial libraries Blood, March 15, 2001; 97(6): 1776 - 1786. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Baker, R. V. Turner, S. J. Gagnon, D. C. Wiley, and W. E. Biddison Identification of a Crucial Energetic Footprint on the {alpha}1 Helix of Human Histocompatibility Leukocyte Antigen (Hla)-A2 That Provides Functional Interactions for Recognition by Tax Peptide/Hla-A2-Specific T Cell Receptors J. Exp. Med., March 5, 2001; 193(5): 551 - 562. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Garcia-Peydro, A. Paradela, J. P. Albar, and J. A. L. d. Castro Antagonism of Direct Alloreactivity of an HLA-B27-Specific CTL Clone by Altered Peptide Ligands of Its Natural Epitope J. Immunol., November 15, 2000; 165(10): 5680 - 5685. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Kalergis and S. G. Nathenson Altered Peptide Ligand-Mediated TCR Antagonism Can Be Modulated by a Change in a Single Amino Acid Residue Within the CDR3{beta} of an MHC Class I-Restricted TCR J. Immunol., July 1, 2000; 165(1): 280 - 285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Gagne, S. Brouard, M. Giral, F. Sebille, A. Moreau, M. Guillet, J.-D. Bignon, B.-M. Imbert, M.-C. Cuturi, and J.-P. Soulillou Highly Altered V{beta} Repertoire of T Cells Infiltrating Long-Term Rejected Kidney Allografts J. Immunol., February 1, 2000; 164(3): 1553 - 1563. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Kalergis, T. Ono, F. Wang, T. P. DiLorenzo, S. Honda, and S. G. Nathenson Single Amino Acid Replacements in an Antigenic Peptide Are Sufficient to Alter the TCR V{beta} Repertoire of the Responding CD8+ Cytotoxic Lymphocyte Population J. Immunol., June 15, 1999; 162(12): 7263 - 7270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Loftus, E. Huseby, P. Gopaul, C. Beeson, and J. Goverman Highly Cross-Reactive T Cell Responses to Myelin Basic Protein Epitopes Reveal a Nonpredictable Form of TCR Degeneracy J. Immunol., June 1, 1999; 162(11): 6451 - 6457. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Hudrisier, J. Riond, O. Burlet-Schiltz, M. G. von Herrath, H. Lewicki, B. Monsarrat, M. B. A. Oldstone, and J. E. Gairin Structural and Functional Identification of Major Histocompatibility Complex Class I-restricted Self-peptides as Naturally Occurring Molecular Mimics of Viral Antigens. POSSIBLE ROLE IN CD8+ T CELL-MEDIATED, VIRUS-INDUCED AUTOIMMUNE DISEASE J. Biol. Chem., May 25, 2001; 276(22): 19396 - 19403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Sharma, J. J. Kuhns, S. Yan, R. H. Friedline, B. Long, R. Tisch, and E. J. Collins Class I Major Histocompatibility Complex Anchor Substitutions Alter the Conformation of T Cell Receptor Contacts J. Biol. Chem., June 8, 2001; 276(24): 21443 - 21449. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
