Highly Cross-Reactive T Cell Responses to
Myelin Basic Protein Epitopes Reveal a
Nonpredictable Form of TCR Degeneracy
This information is current as
of June 16, 2017.
Christine Loftus, Eric Huseby, Priya Gopaul, Craig Beeson
and Joan Goverman
J Immunol 1999; 162:6451-6457; ;
http://www.jimmunol.org/content/162/11/6451
Subscription
Permissions
Email Alerts
This article cites 44 articles, 21 of which you can access for free at:
http://www.jimmunol.org/content/162/11/6451.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 1999 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
References
Highly Cross-Reactive T Cell Responses to Myelin Basic
Protein Epitopes Reveal a Nonpredictable Form of TCR
Degeneracy1
Christine Loftus,* Eric Huseby,† Priya Gopaul,‡ Craig Beeson,* and Joan Goverman2†‡
D41 Th cells recognize peptides that are associated with
MHC class II molecules on the surface of APC. Despite
a high degree of Ag specificity associated with each
TCR, numerous studies have demonstrated flexibility in Ag recognition (1–12). Degeneracy in TCR recognition was first observed in the form of allo-recognition in which T cells selected by
interaction with self-MHC/peptide complexes also respond at a
high frequency to Ag associated with allo-MHC molecules (13–
17). In addition, many TCRs can recognize variants of wild-type
peptides. These “altered peptides” induce T cell responses that
range from full activation to strong antagonism (8, 18 –20). This
ability to transmit different signals depending on the peptide ligand
presumably allows a TCR to trigger positive selection when engaging a ligand in the thymus but trigger full activation of effector
functions when a different ligand is engaged in the periphery (8,
12, 21, 22).
T cell recognition of peptides with more limited sequence homology to the wild-type peptide has also been described (1, 2, 5).
Synthetic peptides with minimal sequence similarity to the natural
ligand are often able to trigger T cell activation if one or two
wild-type amino acid residues that interact with the TCR are retained (1–3, 5). These studies indicated the importance of particular amino acid residues within the bound peptide as TCR contact
residues. A molecular basis for these observations was provided by
crystal structures of peptide/MHC complexes showing residues in
C
Departments of *Chemistry, †Immunology, and ‡Molecular Biotechnology, University of Washington, Seattle, WA 98195
Received for publication January 27, 1999. Accepted for publication March 17, 1999.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
These studies were supported by National Science Foundation Grant MCB-9722374
(to C.B.) and National Institutes of Health Grant R01 NS35126 (to J.G.). J.G. is
supported in part by a Junior Faculty Award (2080-A2) from the National Multiple
Sclerosis Society, and E.H. is supported by a National Institutes of Health Training
Grant CA09537-13.
2
Address correspondence and reprint requests to Dr. Joan Goverman, Department of
Molecular Biotechnology, Box 357650, University of Washington, Seattle, WA
98195. E-mail address: [email protected]
Copyright © 1999 by The American Association of Immunologists
particular positions in the bound peptide available for interaction
with the TCR because they were pointing directly up from the
MHC groove (23). The involvement of these residues in TCR
binding has been further confirmed by crytallographic studies of
peptide/MHC/TCR complexes (24 –28).
TCR degeneracy has also been demonstrated using combinatorial peptide libraries to determine the optimal peptide residues for
TCR recognition at each position (10, 11). These experiments have
suggested that multiple residues could be accommodated at many
positions within a peptide without abolishing TCR recognition.
Furthermore, optimal residues for stimulation were defined for every peptide position, and synthetic peptides with all of these optimal residues resulted in the strongest T cell stimulation. In these
systems, a more stimulatory residue at one contact position could
compensate for a less optimal residue in another contact position.
These studies led to a model in which each TCR contact residue in
the bound peptide contributes independently to facilitate interaction with the TCR, and the overall strength of interaction is the
sum of the individual contributions (10, 11).
Our interest in TCR degeneracy arose from studies of T cells
specific for myelin basic protein (MBP).3 We previously identified
T cells present in MBP-deficient mice that are tolerized by the
endogenous expression of MBP in wild-type mice (29). A large
portion of these T cells exhibited a dual specificity for two adjacent
but nonoverlapping epitopes within MBP. Sequence analysis of
TCR V genes demonstrated that these cross-reactive T cells exhibited a very diverse set of Ag-specific receptors. To understand
how two distinct epitopes bound to the same MHC class II molecule could stimulate a large set of cross-reactive TCRs, we characterized the core peptides that formed these epitopes and identified the residues in each peptide that are accessible to the TCR. We
report here that the TCR contacts in the two cross-reactive epitopes
exhibit no structural similarity to each other. Dissimilarity between
the antigenic surfaces is emphasized by the observation that exchanging an important TCR contact residue from one epitope with
the peptide residue in the analogous position in the other epitope
3
Abbreviation used in this paper: MBP, myelin basic protein.
0022-1767/99/$02.00
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
We identified two nonoverlapping epitopes in myelin basic protein presented by I-Au that are responsible for mediating tolerance
induction to this self-Ag. A large number of T cells expressing diverse TCRs are strongly cross-reactive to both epitopes. Surprisingly, the TCR contact residues in each peptide are highly dissimilar. Furthermore, functional TCR contacts cannot be
interchanged between the two epitopes, indicating that the TCR contacts in each peptide can only be recognized within the context
of the other amino acids present in that peptide’s sequence. This observation indicates that both buried and exposed residues of
each peptide contribute to the sculpting of completely distinct antigenic surfaces. We propose that the cross-reactive TCRs adopt
mutually exclusive conformations to recognize these dissimilar epitopes, adding a new dimension to TCR degeneracy. This unpredictable TCR plasticity indicates that using just the TCR contacts on a single epitope to define other cross-reactive peptides
will identify only a subset of the complete repertoire of cross-reactive epitopes. The Journal of Immunology, 1999, 162: 6451–
6457.
6452
NONPREDICTABLE T CELL DEGENERACY
abolishes T cell recognition. These data indicate that the TCR
contact residues in these epitopes are not recognized independently. Instead, recognition of the TCR contact residues is context
dependent and, for these two epitopes, is mediated by mutually
exclusive conformations of individual TCRs. The ability of TCRs
to exhibit this type of degeneracy greatly limits the ability to identify all possible cross-reactive epitopes based on the sequence of a
single epitope.
Materials and Methods
Generation of T cell hybridomas
Hybridomas were generated as previously described (29). Briefly, T cells
were isolated from the draining lymph nodes of H-2u MBP2/2 (shiverer)
mice (29) previously immunized with 75 mg murine MBP and stimulated
for 72 h with 30 mM MBP131–150 before fusion. MBP was purified from
mouse brains (Pel-Freeze Biologicals, Rogers, AR) (30). The hybridomas
were screened for positive responses to both MBP and MBP131–150 using
HT-2 cells to detect IL-2 production before cloning by limiting dilution.
Peptide synthesis
Dissociation rates of peptide/MHC complexes
I-Au protein was isolated as described previously (31). In brief, cells expressing I-Au were lysed and passed over a lentil lectin column that was
subsequently eluted with methyl mannoside onto an affinity column (10.3.6
Ab). The protein was eluted from the Ab column with Na2CO3, pH 11.5,
purity was assessed with silver-stained SDS-PAGE, and concentrations
were determined with a micro bicinchoninic acid assay (Pierce, Rockford,
IL). To measure the rate of dissociation of fluorescein-labeled peptides, a
solution of I-Au protein and an excess of peptide were incubated at pH 5.3
at 37°C for 24 or 48 h. Unbound peptide was then removed by size exclusion (Sephadex G50-SF) at 4°C. The reaction mixture was separated by
high performance size exclusion chromatography using a 60- or 30-cm by
7.5-mm TSK3000SW column (Toso Haas, Montgomeryville, PA) and a
fluorescence detector. At the beginning of the dissociation, the initial
amount of labeled peptide bound to the MHC was measured as the peak
height of the peptide/MHC fraction. After subsequent incubations at 37°C,
the relative peak height at each time was used as a measure of peptide still
bound to the protein. Half-times of dissociation were obtained from single
exponential fits to the dissociation data.
T cell stimulation
T cell responses to Ag were assessed by measuring the amount of IL-2
present in culture supernatants after incubating 1 3 105 T cell hybridomas
with 5 3 105 irradiated spleen cells and either 20 mM peptide, 3 mM MBP,
or no Ag. Incubations were conducted in duplicate or triplicate for 48 h in
96-well round-bottom plates in a total volume of 200 ml growth media
containing DMEM supplemented with 10% FCS. Supernatant (50 ml/well)
was transferred to another 96-well plate and frozen at 280°C. Relative
amounts of IL-2 in the supernatants were determined by adding IL-2-dependent HT-2 indicator cells (1 3 104/well) and measuring the proliferation of the cells after 24 h by adding [3H]thymidine for the last 8 h of
incubation.
Results
Specificity of T cell hybridomas
Previous studies of T cell hybridomas specific for MBP121–140
divided the hybridomas into three groups based on their fine specificity for Ag (29). The first group recognized only the MBP121–
140 peptide, the second group recognized MBP121–140 and the
nested peptide MBP126 –140, and the third group recognized
MBP121–140, MPB126 –140, and the partly overlapping peptide
FIGURE 1. Predicted alignment of the two MBP epitopes. A, The backbone conformation of a generic polyalanine peptide bound to a MHC class
II protein with the P1–P9 positions labeled. Coordinates were derived from
the crystal structure of OVA323–339 peptide bound to I-Ad (32). B, Predicted alignments for the two epitopes in which P1, P4, P6, and P9 positions are represented with downward arrows and P2, P5, and P8 positions
are represented with upwards arrows, respectively.
MBP131–150. However, the group III hybridomas did not respond
to MBP131–140, the only region of overlap between MBP121–
140 and MBP131–150. Thus, group III hybridomas exhibit an unusual specificity in that these T cells could be stimulated by two
distinct epitopes, one within MBP121–140 and the other within
MBP131–150. To confirm that the dual specificity for the two
epitopes was encoded by individual TCRs, the TCR a- and
b-chain genes from two different group III hybridomas were
cloned and independently transfected into a recipient hybridoma
lacking TCR expression. In both cases, expression of the transfected genes from a single TCR was sufficient to confer strong
recognition of both MBP121–140 and MBP131–150 (data not
shown).
Prediction of MHC-binding epitopes
To predict how sequences within this region of MBP might bind to
I-Au, we evaluated existing structural information on the I-Au molecule. Crystal structures of other murine MHC class II molecules
complexed with antigenic peptides have revealed a highly conserved backbone conformation of bound peptides (23, 32, 33).
Thus, for each residue in a peptide bound to a MHC class II molecule, it is possible to predict whether the side chain is pointing
directly down into the binding groove, up toward the TCR, or
somewhere in between (Fig. 1A).
Although a crystal structure has not been determined for I-Au,
the properties of the binding groove have been well described by
peptide binding studies and molecular modeling (31, 34 –37). The
side chain of the residue at the P1 position (i.e., the P1 side chain)
points down into a binding pocket of I-Au that is relatively large
and hydrophobic. The P2 side chain points up and out of the binding groove and is the first potential T cell contact. The P3 side
chain occupies a very shallow, solvent-exposed pocket, and the P4
side chain points down into a pocket that is smaller than the P9
pocket. The major T cell contact residue of the I-Au motif is generally found at the P5 position, where the side chain is pointed
directly up and out of the groove. The P6 side chain points down
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
Peptides were synthesized with standard Fast F-moc chemistry on an Applied Biosystems 431A peptide synthesizer (Foster City, CA) and labeled
on the N terminus with the N-hydroxysuccinimidyl ester of 5(6)-carboxyfluorescein before cleavage. Cleavage from the resin was achieved with
85% trifluoracetic acid, 10% water, and 5% thioanisole, and crude peptides
were then purified by reverse-phase HPLC (acetonitrile/water gradient with
0.1% trifluoracetic acid). Identity of the purified peptides was confirmed by
electrospray mass spectrometry, and concentrations of peptide solutions
were obtained through absorbance measurements of the fluorescein label in
pH 8.9 buffer at 495 nm.
The Journal of Immunology
Verification of predicted binding epitopes
Peptides of varying lengths within MBP121–150 and sequence
variations of the two core peptides were synthesized and tested for
their ability to bind to I-Au and stimulate T cells. The combination
of the T cell stimulation data with measurements of peptide/MHC
stability (Table I) was used to determine the extent to which a
specific peptide side chain interacts with either the MHC binding
groove, the TCR, or both. The half-time (t1/2) of dissociation of
each peptide/I-Au complex was measured by incubating fluorescein-labeled peptide with detergent soluble I-Au and measuring the
dissociation rate by chromatography. T cell stimulation was assessed by measuring IL-2 production in response to Ag stimulation. Response to each peptide was tested using panels of six or
nine different T cell hybridomas. The number of hybridomas that
responded with a weak (0 –10%), moderate (10 –50%), or complete
(50 –100%) response relative to the core peptide is reported.
Table I. Peptide variants of MBP121–150 used in study
Stimulationb
Peptide
t1/2 (h)a
121–140
124–140
125–140
126–140
127–140
128–140
129–140
130–140
126–140, G126I
126–140, A128T
126–140, A128I
126–140, S129E
126–140, D130A
126–140, Y131E
125–135 (core 1)
125–135, R127A
125–135, D130A
125–135, S133E
125–135, S133A
132–144
136–148
134–146
140–150
136–146 (core 2)
136–146, G137I
136–146, F138A
136–146, F138Y
136–146, K139A
136–146, A141D
136–146, Y142E
136–146, D143A
136–146, A144Q
136–146, A144D
136–146, Q145A
200
213
210
76
8.4
5.2
,1
,1
125
5
,1
3
28
,1
270
24
40
270
270
2
20
180
,1
180
180
170
180
5
300
,1
15
,1
5
,1
P1
P3
P3
P4
P5
P6
P2
P5
P8
P8
P1
P2
P2
P3
P5
P6
P7
P8
P8
P9
0–10%
10–50%
50–100%
0/8
4/8
0/8
2/8
8/8
2/8
0/9
6/9
7/9
4/9
0/6
0/4
0/9
0/9
1/9
1/9
0/6
1/4
9/9
3/9
1/9
4/9
6/6
3/4
2/7
0/7
5/7
0/9
3/9
0/9
7/9
0/9
7/9
8/9
9/9
6/9
9/9
4/9
0/9
3/9
2/9
2/9
5/9
1/9
0/9
0/9
0/9
0/9
1/9
9/9
3/9
7/9
0/9
4/9
1/9
1/9
0/9
3/9
0/9
4/9
a
Half-times (t1/2) of dissociation from I-Au were measured by using high performance size exclusion chromatography to quantitate the amount of fluorescein-labeled
peptide bound to I-Au after serial incubations at 37°C.
b
T cell stimulation of each peptide is represented as the fraction of hybridomas
that responded within a range defined for the native core.
Length variation peptides define core epitopes
A properly aligned 11-mer peptide should exhibit the most stable
binding because it encompasses most of the MHC binding interactions. The peptide within MBP121–140 that formed the most
stable complex with I-Au is the MBP125–135 peptide (Fig. 2),
strongly supporting the hypothesis that the core epitope consists of
these 11 residues. Extension of the peptide to include residues
outside of the core region did not increase binding and, as the core
was truncated, binding decreased (Table I and Fig. 2). In addition
to exhibiting the most stable binding to I-Au, MBP125–135 stimulates all T cell hybridomas in the panel as well as MBP121–140
(Table I). The truncated MBP127–140 peptide stimulated all T
cells despite much weaker binding (t1/2 5 5 h), while further truncation of the P2 residue, a putative TCR contact, eliminated much
of the T cell stimulation (Table I).
The MBP136 –146 peptide also exhibited a very long half-life,
and extensions at either end of MBP136 –146 did not increase the
binding stability. Thus, every residue required for the maximal
stability is contained in MBP136 –146 (Fig. 2). Binding of the
MBP132–144 peptide is greatly diminished (t1/2 5 2 h) due to the
loss of the P9 residue. However, this level of binding is sufficient
for the peptide to fully stimulate many of the T cells (Table I). The
MBP132–144 peptide does retain all of the residues predicted to be
accessible to the TCR. We had also considered MBP138 –148 as a
possible second core epitope because it would align D143 with the
D130 at the P5 position of the first epitope. However, this align-
ment would position the Y142 side chain into the very shallow P4
pocket and the K139 side chain into the nonpolar P1 pocket. It was
also found that extending MBP136 –146 to MBP136 –148, a peptide that includes the P8 and P9 residues of this alternate alignment, did not improve MHC binding.
Responses to core peptides reveal extensive TCR cross-reactivity
Analyses of T cell responses to the core peptides redefined the Ag
specificity of the hybridomas that respond to this region of MBP.
All of the hybridomas that were tested from group II (10/10) and
group III (12/12) exhibited cross-reactivity for MBP125–135 and
MBP136 –146. The cross-reactive response of the group II hybridomas to MBP136 –146 was unexpected because these T cells were
previously distinguished from group III by their inability to respond to the longer MBP131–150 peptide that contains this core
epitope. Our previous studies showed that the small number of
MBP121–150-specific T cells that escape tolerance in wild-type
mice consist of both group I and group II T cells. Therefore, the
ability of group II T cells to recognize the two core epitopes in this
region indicates that T cells exhibiting cross-reactivity are found in
the periphery of wild-type mice.
Five group I hybridomas that only recognized MBP121–140
peptide were also tested for cross-reactivity to the two core peptides. In contrast to group II and group III hybridomas, none of
these hybridomas responded significantly to MBP136 –146. Three
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
into the groove, filling an unusually large, hydrophobic pocket
bounded by I-Au residues Tyr b30 and Phe b11. The binding preferences of the P6 pocket are best illustrated by studies of MBP1–
11/I-Au complexes in which a strong preference for tyrosine at the
P6 position is observed (37). The P7 side chain occupies a shallow,
solvent-exposed pocket situated near the MHC b72 Glu residue, a
potential partner for salt bridge formation. The P8 side chain points
upward and is the final potential T cell contact. The P9 side chain
is positioned in an MHC pocket of moderate size and polarity.
Hypothetical alignments for the two MBP epitopes were proposed based on the structural characteristics of the I-Au binding
groove, the primary sequence of MBP121–150, and our preliminary binding studies (29). To test our predictions, core epitopes
comprised of 11 residues were synthesized because the flanking
P(-1) and P10 residues contribute to peptide/MHC stability by
forming hydrogen bonds between MHC side chains and the peptide backbone (23, 38, 39). The MBP peptides 125–135 and 136 –
146 were predicted to be the core epitopes as shown in Fig. 1B. In
both alignments, a large, aromatic tyrosine side chain fills the P6
pocket, forming a primary binding anchor for the peptide.
6453
6454
NONPREDICTABLE T CELL DEGENERACY
and seven Vb subfamilies were represented among 34 Vb genes.
Furthermore, a diverse set of Ja and Jb gene segments were associated with these V genes (29). A comparison of the CDR3 sequences for a set of Va and Vb genes expressed on group II and
group III hybridomas (Table II) illustrates that there are no obvious
highly conserved structural features in these TCRs that could account for the shared specificity of the T cells.
Single amino acid substitutions that affect MHC binding
group I hybridomas did respond to MBP125–135, and two did not
respond to any peptide tested except MBP121–140. Apparently
recognition by some of the group I T cells requires additional
residues at either the amino- or carboxy-terminal ends of
MBP125–135. Representative responses of group I, II, and III hybridomas are shown in Fig. 3. The T cell hybridomas used for the
data in Table I were comprised of both groups II and III.
Hybridomas from all specificity groups respond to MBP121–
140, consistent with the fact that these hybridomas were obtained
by stimulating MBP-primed T cells with MBP121–140 in vitro
before fusion. To determine whether T cells specific for MBP131–
150 exist that do not cross-react with MBP121–140, additional
hybridomas were generated by immunizing MBP-deficient animals with MBP and stimulating cells from the draining lymph
nodes with MBP131–150 before fusion. Twelve independent hybridomas were obtained with this protocol, and all responded to
both MBP121–140 and MBP131–150. Thus, there do not appear to
be T cells that can recognize MBP131–150 without also recognizing MBP121–140.
A diverse set of TCRs exhibit cross-reactivity to MBP125–135
and MBP136 –146
The degenerate recognition of MBP125–135 and MBP136 –146
was observed for more than 30 different T cell hybridomas (data
not shown). Remarkably, the repertoire of Va and Vb genes expressed on these hybridomas is quite diverse. Our previous studies
demonstrated that nine different Va subfamilies were represented
among 29 Va genes expressed on group II and group III T cells,
Single amino acid substitutions that affect T cell recognition
Substitution of a peptide residue that strongly affects T cell stimulation while retaining sufficient MHC binding suggests that the
peptide side chain at that position is directed outside of the groove
and available for interaction with a TCR. However, the accessibility of a residue does not guarantee that it is essential for recognition by all T cells that recognize this epitope. The diversity in
group II and III TCRs suggests that there may be a corresponding
diversity in the TCR residues that interact with these peptides. To
assure that mutations of a TCR contact not detected by one hybridoma would likely be detected by another, a panel of nine hybridomas differing in V gene usage was used to assess the effects
of substitutions on TCR recognition.
All of the hybridomas were extremely sensitive to substitution
of the P5 residue on both core peptides, while mutations of the P2
and P8 residues had differential effects (Fig. 4). Of particular importance, substituted peptides in which the wild-type P5 residues
were interchanged between the first and second epitope (i.e.,
D130A for the first and A141D for the second) were not recognized by any T cell hybridomas. Therefore, even though each
cross-reactive TCR can recognize both an aspartic acid and an
alanine at the P5 position, the aspartic acid can only be recognized
in the context of MBP125–135 and the alanine only in the context
of MBP136 –146.
Discussion
In the studies described here, we investigated the structural basis
for the degenerate TCR recognition of two MBP epitopes. Our
analyses using length-variation peptides defined two core epitopes
within the MBP121–150 region, MBP125–135 and MBP136 –146.
Alignment of these peptides in the MHC groove predicted the
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
FIGURE 2. Definition of the core epitopes using peptides with extensions at the N and C termini. A, Within MBP121–140, the 125–135 peptide
was most stable. B, Within MBP131–150, the 136 –146 peptide was
most stable. Dissociation half-times (t1/2) were measured as described
in Table I.
The function of specific residues in the two epitopes was studied
using peptides containing single amino acid substitutions. Substitutions in each epitope are divided into those that essentially affect
only MHC binding and those that retained sufficient MHC binding
but primarily lost T cell recognition relative to the wild-type sequences. Our predicted alignments for the MBP125–135 and
MBP136 –146 core peptides were predicated on placing a tyrosine
side chain in the P6 pocket (Y131 and Y142, respectively). In
support of this assignment, substitution of these tyrosine residues
with the polar glutamic acid completely eliminated binding for
both peptides (Table I). Although the P1 pocket of I-Au appears to
be fairly large and hydrophobic, both predicted epitopes have a
glycine at the P1 position that would not place a side chain into the
P1 pocket. Replacement of the P1 glycine with a larger nonpolar
residue, such as isoleucine, was expected to be well tolerated and
potentially increase binding. An increase in stability was observed
for G126I in the first core epitope, and wild-type stability was
maintained for G137I in the second core epitope (Table I). In contrast, the P3 and P7 pockets are very shallow and therefore are
expected to be much less tolerant of changes in the side chains.
Substitutions at either the P3 or P7 positions in both core peptides
diminished MHC binding (Table I).
The Journal of Immunology
6455
“topology” of each peptide when bound to its ligand, i.e., which
residues should be primarily involved in MHC binding and which
should be accessible to the TCR. The effects of numerous amino
acid substitutions on MHC binding and T cell recognition were
then used to confirm the predicted alignments for each core peptide. Finally, peptide residues that appeared to be accessible to the
TCR were further analyzed to determine which of these residues
contributed to the specificity of T cell recognition (“functional”
TCR contacts) and which were at the TCR/MHC interface but
were not strictly required for recognition. Similarities between the
functional TCR contact residues in the two epitopes would suggest
that cross-reactivity occurs by a molecular mimicry mechanism
(40, 41), while a lack of similarity would point to an alternate
mechanism.
In MBP125–135, the aspartic acid at the P5 position contributed
strongly to the specific recognition of this epitope. Substitution of
the aspartic acid with an alanine (D130A) had only a minor effect
on MHC binding but eliminated recognition by nearly all T cells.
The arginine residue at the P2 position also appears to be an important functional TCR contact in that only a few T cells tolerated
its mutation to an alanine residue (R127A). The serine at the P8
position behaved differently: the side chain is accessible to the
TCR but does not appear to contribute significantly to recognition.
The loss of stimulation following the replacement of serine with
Table II. Sequences of V gene regions for selected T cell hybridomas
Hybridoma
Va
Variable
n
Joining
Ja
Vb
Variable
nDn
1.G2.C2 (II)
1.G8.C3 (II)
2.E2.D6 (II)
2.D10.D3 (II)
1.D7.C1 (II)
2.B2.D5 (II)
2.F6.C4 (III)
2.G11.A5 (III)
2.C11.C5 (III)
1.D9.B2 (III)
Va19
Va11
Va19
Va8
Va2
Va19
Va19
Va2
Va2
Va2
YFCAA
YFCA
YFCAA
YCAL
YFCAA
YFCAA
YFCAA
YFCAA
YFCAA
YFCAA
SM
AV
A
N
S
SM
SM
SH
SAP
I
DSNYQLIW
SNYNVLYF
DTNAYKVIF
TGANYGKLTF
DTNAYKVIF
TNAYKVIF
DSNYQLIW
NNNAPRF
GGYKVVF
DTNAYKVIF
Ja26
Ja16
Ja23
Ja44
Ja23
Ja23
Ja26
Ja35
Ja23
Ja23
Vb2
Vb3
Vb4
Vb8.2
Vb8.3
Vb8.3
Vb2
Vb2
Vb4
Vb8.2
CTCSA
CXS
CASSQ
CASGD
CASS
CASS
CTFSA
CTFS
CASS
CASGD
AGTD
RQGWDA
RDRD
AKGTGH
GLDRDS
PPGTDK
GGVDD
ETAGR
DRGAD
LGQT
Joining
NERLF
EQF
NQAPLF
TGQLY
DTQY
NTLY
ERXF
DTQY
NQAPLF
NERLF
Jb
Jb1.4
Jb2.1
Jb1.5
Jb2.2
Jb2.5
Jb2.4
Jb1.4
Jb2.5
Jb1.5
Jb1.4
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
FIGURE 3. Cross-reactivity of MBP-specific T cells for the MBP125–135 and 136 –146 core epitopes. Individual T cell hybridomas previously defined
as group I, II, and III (29) were tested for their ability to recognize the core peptides as well as the longer MBP peptides previously used to define Ag fine
specificity. Top panels show the responses of two different group I hybridomas representative of those that do (left) and those that do not (right) recognize
MBP125–135. Bottom panels show representative responses of a group II (left) and group III (right) hybridoma. IL-2 production by the hybridomas in
response to stimulation with different peptides was measured as proliferation of HT-2 indicator cells.
6456
FIGURE 4. Effects of substitution of residues predicted to be accessible
to the TCR. Most of the substitutions abolished T cell recognition even
though MHC binding was sufficient for complete activation. Dissociation
half-times (t1/2) and the number of hybridomas that responded in each of
the three stimulation categories are listed as presented in Table I.
TCR residues have substantially shifted when bound to the second
epitope. In this alternate TCR conformation, binding to an aspartic
acid at the P5 position is no longer permitted. An aspartic acid at
the P7 position is required for binding to the second epitope. This
raised the possibility that the TCR is directed into position by an
aspartic acid residue whose position is shifted in the second
epitope. However, specific recognition by the cross-reactive TCRs
is not simply dependent on the presence of an aspartic acid at
either the P5 or P7 position. If this were the case, then T cells
would have recognized the second epitope substituted with an aspartic acid at the P5 position (A141D).
Our experiments suggest that the TCR conformation adopted for
recognition of the first epitope structurally excludes the recognition of the second epitope, and vice versa. Thus, not only are
distinct conformations of the TCR required for binding each of the
two epitopes, but these TCR conformations are mutually exclusive. Furthermore, the observation that over 30 diverse T cells are
cross-reactive for the two epitopes suggests that this ability of a
TCR to adopt at least two alternate, mutually exclusive conformations is not a rare event.
Recent studies have proposed several mechanisms by which degeneracy of TCR recognition may occur. It has been well established that peptide residues buried within the binding groove of the
MHC can have strong influences on T cell recognition (18, 42).
Other experiments have shown that mutation of one TCR contact
residue on a peptide can alter the recognition of a second residue
in that peptide (43). While both of these observations could be
explained by only a change of the peptide conformation, experiments conducted by Ono et al. (44) suggest that substitution of a
peptide residue buried in the MHC can have consequences that are
more global in nature. Conservative mutations of TCR contact
residues in a viral peptide were found to alter the pattern of CTL
recognition of MHC residues. It was concluded that alterations in
the TCR conformation allow it to bind related ligands. The striking
feature of these results is that many of the TCR contacts on the
MHC protein were remote from the substituted peptide residue,
indicating that changes in the peptide structure can induce conformational changes throughout the MHC surface.
Our results demonstrate that when MBP125–135 and MBP136 –
146 separately bind to I-Au, two distinct antigenic surfaces are
generated that are uniquely defined by the peptide sequences. Our
conclusion that TCR recognition of a peptide residue might occur
only in the context of one peptide sequence and be prohibited in a
different sequence is in direct contrast with the suggestion that
TCR contacts function independently of each other (10, 11). The
optimization of TCR contacts independently of each other has
been used to identify cross-reactive epitopes; however, this is only
successful when the overall conformation of the TCR bound to
each epitope is similar. If, as observed here, binding of two crossreactive peptides to the same MHC molecule produces unrelated
antigenic surfaces, then a TCR bound to one surface may be required to adopt a conformation that excludes the ability to recognize the other surface. Thus, MBP125–135 and MBP136 –146
would never have been predicted to be cross-reactive by simply
comparing the functionally important TCR contacts in the two
epitopes. The mutually exclusive contexts in which the two MBP
epitopes are recognized indicate that TCR degeneracy is likely to
be much broader and less predictable than previously believed.
Therefore, an attempt to predict potential cross-reactive epitopes
based only on the analysis of T cell contact residues in a particular
peptide sequence will be intrinsically limited in scope.
Our studies do not imply that the induction of tolerance in T
cells specific for MBP121–150 depends upon their cross-reactivity
for two distinct epitopes within this region. It is possible that only
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
glutamic acid (S133E) confirms that this residue is at the TCR/
MHC interface. However, substitution of the serine with an alanine
(S133A) had little effect on recognition, indicating that the serine
side chain itself is not strictly required.
Defining the functional TCR contacts in the second epitope was
less straightforward. Mutation of the phenylalanine at the P2
position produced similar effects to those observed for the serine at
the P8 position in MBP125–135. Substitution of this phenylalanine
with a tyrosine (F138Y) had no effect on MHC binding but abolished recognition by most T cells. This disruption of recognition
due to insertion of an oxygen atom at this position strongly suggests that this residue lies at the TCR/MHC interface. However,
substitution of the phenylalanine with the much smaller alanine
residue (F138A) did not affect either MHC binding or T cell stimulation. Therefore, the affinity between the TCR and MHC ligand
does not strictly require interaction with the phenylalanine. At the
P5 position, mutation of the alanine to an aspartic acid (A141D)
did not diminish MHC binding but eliminated recognition by most
hybridomas. Although this result demonstrates that the specificity
of peptide recognition is dependent on the alanine, binding of the
methyl side chain of alanine to a TCR pocket is unlikely to contribute substantially to binding affinity. Similarly, substitution of
the alanine residue at the P8 position to an aspartic acid abolished
all T cell recognition. The same reservation in attributing significant TCR binding affinity to the methyl side chain of an alanine
applies to residues at both the P5 and P8 position.
Because none of the residues that we had so far evaluated in
MBP136 –146 seemed well suited as functional TCR contacts, the
residues at the P3 and P7 positions of this epitope were also studied. Residues at these positions are potential TCR contacts in that
they reside in shallow, solvent-exposed pockets. Substitution of
the aspartic acid at the P7 position with an alanine (D143A) retained sufficient MHC binding but abolished recognition by all T
cells. Thus, this aspartic acid appears to be a functional TCR contact. In contrast, substitution of the lysine with an alanine at the P3
position (K139A) had little effect on T cell recognition. Together,
these results indicate that the functional TCR contacts in the second epitope are distinct from those in the first epitope.
The observation that the peptide residues most important for
specificity of T cell recognition differ between the two peptides
rules out the mechanism of molecular mimicry. Instead, our results
suggest a mechanism for recognition by which the contribution of
each TCR contact is dependent upon the context of the entire peptide sequence. This is best illustrated by the fact that an aspartic
acid residue at the P5 position is necessary for TCR recognition of
the first epitope, while an aspartic acid at the equivalent position on
the second epitope prohibits T cell recognition. The inability to
tolerate an exchange of a functional TCR contact suggests that
NONPREDICTABLE T CELL DEGENERACY
The Journal of Immunology
Acknowledgments
We are grateful to Dr. T. Brabb and A. Perchelett for critical reading of the
manuscript.
References
1. Bhardwaj, V., V. Kumar, H. M. Geysen, and E. E. Sercarz. 1993. Degenerate
recognition of a dissimilar antigenic peptide by myelin basic protein-reactive T
cells. J. Immunol. 151:5000.
2. Evavold, B. D., J. Sloan-Lancaster, K. J. Wilson, J. B. Rothbard, and P. M. Allen.
1995. Specific T cell recognition of minimally homologous peptides: evidence for
multiple endogenous ligands. Immunity. 2:655.
3. Nanda, N. K., K. K. Arzoo, H. M. Geysen, A. Sette, and E. E. Sercaz. 1995.
Recognition of multiple peptide cores by a single T cell receptor. J. Exp. Med.
182:531.
4. Quaratino, S., D. J. Thorpe, P. J. Travers, and M. Londei. 1995. Similar antigenic
surfaces, rather than sequence homology, dictate T cell epitope molecular mimicry. Proc. Natl. Acad. Sci. USA 92:10398.
5. Wucherpfennig, K. W., and J. L. Strominger. 1995. Molecular mimicry in T
cell-mediated autoimmunity: viral peptides activate human T cell clones specific
for myelin basic protein. Cell 80:695.
6. Brock, R., K.-H. Wiesmuller, G. Jung, and P. Walden. 1996. Molecular basis for
the recognition of two structurally different major histocompatibility complex/
peptide complexes by a single T cell receptor. Proc. Natl. Acad. Sci. USA 93:
13108.
7. Kersh, G. J., and P. M. Allen. 1996. Structural basis for T cell recognition of
altered peptide ligands: a single T cell receptor can productively recognize a large
continuum of related ligands. J. Exp. Med. 184:1259.
8. Kersch, G. J., and P. M. Allen. 1996. Essential flexibility in the T cell recognition
of antigen. Nature 380:495.
9. Tallquist, M. D., T. J. Yun, and L. R. Pease. 1996. A single T cell receptor
recognizes structurally distinct MHC/peptide complexes with high specificity.
J. Exp. Med. 184:1017.
10. Hemmer, B., B. T. Fleckenstein, M. Vergelli, G. Jung, H. McFarland, R. Martin,
and K.-H. Wiesmuller. 1997. Identification of high potency microbial and self
ligands for a human autoreactive class II-restricted T cell clone. J. Exp. Med.
185:1651.
11. Hemmer, B., M. Vergelli, B. Gran, N. Ling, P. Conlon, C. Pinilla, R. Houghten,
H. F. McFarland, and R. Martin. 1998. Predictable TCR antigen recognition
based on peptide scans leads to the identification of agonist ligands with no
sequence homology. J. Immunol. 160:3631.
12. Ignatowicz, L., W. Rees, R. Pacholczyk, H. Ignatowicz, E. Kushnir, J. Kappler,
and P. Marrack. 1997. T cells can be activated by peptides that are unrelated in
sequence to their selecting peptide. Immunity 7:179.
13. Udaka, K. K., H. Wiesmuller, S. Kienle, G. Jung, and P. Walden. 1996. Self
MHC-restricted peptides recognized by an alloreactive T lymphocyte clone.
J. Immunol. 157:670.
14. Bevan, M. J. 1977. In a radiation chimaera, host H-2 antigens determine immune
responsiveness of donor cytotoxic cells. Nature 269:417.
15. Zinkernagel, R. M., G. N. Callahan, A. Althage, S. Cooper, P. A. Klein, and
J. Klein. 1978. On the thymus in the differentiation of “H-2 self-recognition” by
T cells: evidence for dual recognition? J. Exp. Med. 147:882.
16. Zinkernagel, R. M., G. N. Callahan, A. Althage, S. Cooper, W. Streilen, and
J. Klein. 1978. The lymphoreticular system in triggering virus plus self-specific
cytotoxic T cells: evidence for T help. J. Exp. Med. 147:897.
17. Zinkernagel, R. M., G. N. Callahan, J. Klein, and G. Dennert. 1978. Cytotoxic T
cells learn specificity for self H-2 during differentiation in the thymus. Nature
271:251.
18. Sloan-Lancaster, J., and P. M. Allen. 1996. Altered peptide ligand-induced partial
T cell activation: molecular mechanisms and role in T cell biology. Annu. Rev.
Immunol. 14:1.
19. Nichelson, L. B., and V. K. Kuchroo. 1997. T cell recognition of self and altered
self antigens. Crit. Rev. Immunol. 17:449.
20. Ruppert, J., J. Alexander, K. Snoke, M. Coggeshal, E. Herbert, D. McKenzie,
H. Grey, and A. Sette. 1993. Effect of T-cell receptor antagonism on interaction
between T cells and antigen presenting cells and on signalling events. Proc. Natl.
Acad. Sci. USA 90:2671.
21. Hogquist, K. A., M. A. Gavin, and M. J. Bevan. 1993. Positive selection of CD81
T cells induced by major histocompatibility complex binding peptides in fetal
thymic organ culture. J. Exp. Med. 177:1469.
22. Hogquist, K. A., S. C. Jameson, W. R. Heath, J. L. Howard, and F. R. Carbone.
1994. T cell receptor antagonist peptides induce positive selection. Cell 76:17.
23. Jardetzky, T. S., J. H. Brown, J. C. Gorga, L. J. Stern, R. G. Urban,
J. L. Strominger, and 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.
24. Garboczi, D. N., P. Ghosh, U. Utz, Q. R. Fan, W. E. Biddison, and D. C. Wiley.
1996. Structure of the complex between human T cell receptor, viral peptide, and
HLA-A2. Nature 384:134.
25. Garcia, K. C., M. Degano, R. L. Stanfield, A. Brunmark, M. R. Jackson,
P. A. Peterson, L. Teyton, and I. A. Wilson. 1996. An ab T cell receptor structure
at 2.5 A and its orientation in the TCR-MHC complex. Science 274:209.
26. Ding, Y.-H., K. J. Smith, D. N. Garboczi, U. Utz, W. E. Biddison, and
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
9:403.
27. Garcia, K. C., M. Degano, L. R. Pease, M. Huang, P. A. Peterson, L. Teyton, and
I. A. Wilson. 1998. Structural basis of plasticity in T cell receptor recognition of
a self peptide-MHC antigen. Science 279:1166.
28. 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, and J.-H. Wang. 1998. Identification of a common docking topology with substantial variation among different
TCR-peptide-MHC complexes. Curr. Biol. 8:409.
29. Harrington, C. J., A. Paez, T. Hunkapiller, V. Mannikko, T. Brabb, M. Ahearn,
C. Beeson, and J. Goverman. 1998. Differential tolerance is induced in T cells
recognizing distinct epitopes of myelin basic protein. Immunity 8:571.
30. Norton, W. T., and S. E. Poduslo. 1973. Myelination in rat brain: method of
myelin isolation. J. Neurochem. 21:749.
31. Mason, K., D. W. Denney, Jr., and H. M. McConnell. 1995. Kinetics of the
reaction of a myelin basic protein peptide with soluble I-Au. Biochemistry 34:
14874.
32. Scott, C. A., P. A. Peterson, L. Teyton, and I. A. Wilson. 1998. Crystal structure
of two I-Ad-peptide complexes reveal that high affinity can be achieved without
large anchor residues. Immunity 8:319.
33. Fremont, D. H., D. Monnaie, C. A. Nelson, W. A. Hendrickson, and
E. R. Unanue. 1998. Crystal structure of I-Ak in complex with a dominant epitope
of lysozyme. Immunity 8:305.
34. Liang, M. N., C. Beeson, K. Mason, and H. M. McConnell. 1995. Kinetics of the
reactions between the invariant chain (85–99) peptide and proteins of the murine
class II MHC. Int. Immunol. 7:1397.
35. Mason, K., D. W. Denney, Jr., and H. M. McConnell. 1995. Myelin basic protein
peptide complexes with the class II MHC molecules I-Au and I-Ak form and
dissociate rapidly at neutral pH. J. Immunol. 154:5216.
36. Liang, M. N., C. Lee, Y. Xia, and H. M. McConnell. 1996. Molecular modeling
and design of invariant chain peptides with altered dissociation kinetics from
class II MHC. Biochemistry 35:12734.
37. Lee, C., M. N. Liang, K. M. Tate, J. D. Rabinowitz, C. Beeson, P. P. Jones, and
H. M. McConnell. 1998. Evidence that the autoimmune antigen myelin basic
protein (MBP) Ac1–9 binds towards one end of the major histocompatability
(MHC) cleft. J. Exp. Med. 9:1505.
38. Sette, A., S. Buus, S. Colon, J. A. Smith, C. Miles, and H. M. Grey. 1987.
Structural characteristics of an antigen required for its interaction with Ia and
recognition by T cells. Nature 328:395.
39. Rudensky, A. Y., P. Preston-Hurlburt, B. K. Al-Ramadi, J. Rothbard, and
C. A. Janeway, Jr. 1992. Truncation variants of peptides isolated from MHC class
II molecules suggest sequence motifs. Nature 358:429.
40. Olstone, M. B. A. 1987. Molecular mimicry and autoimmune disease. Cell 50:
819.
41. Theofilopoulos, A. N. 1995. The basis of autoimmunity. I. Mechanisms of aberrant self-recognition. Immunol. Today 16:90.
42. Dyall, R., D. H. Fremont, S. C. Jameson, and J. Nikolic-Zugic. 1996. T cell
receptor (TCR) recognition of MHC class I variants: intermolecular second-site
reversion provides evidence for peptide/MHC conformational variation. J. Exp.
Med. 184:253.
43. Ausubel, L. J., C. K. Kwan, A. Sette, V. Vuchroo, and D. A. Hafler. 1996.
Complementary mutations in an antigenic peptide allow for crossreactivity of
autoreactive T-cell clones. Proc. Natl. Acad. Sci. USA 93:15317.
44. Ono, T., T. P. DiLorenzo, F. Wang, A. M. Kalergis, and S. G. Nathenson. 1998.
Alterations in TCR-MHC contacts subsequent to cross-recognition of class I
MHC and singly substituted variants. J. Immunol. 161:5454.
Downloaded from http://www.jimmunol.org/ by guest on June 16, 2017
one of these epitopes is actually presented in vivo and is responsible for mediating tolerance to the MBP protein. In this regard, it
is interesting to note that all hybridomas that were obtained by
stimulating T cells in vitro with MBP131–150 before fusion responded to both MBP121–140 and MBP131–150. We also have no
evidence to suggest that this form of degeneracy is limited to
tolerogenic epitopes. In fact, the degeneracy exhibited by the T
cells described here was only discovered during our analyses of the
immune response to a self-Ag because the two cross-reactive MBP
epitopes are adjacent to each other in the primary sequence of the
protein. The cross-reactivity of these TCRs was revealed because
the set of synthetic peptides used to define their fine specificity
happen to include nonoverlapping peptides that contained each
epitope. While this close proximity allowed us to detect the crossreactive epitopes, it is not required or even expected that such
cross-reactive epitopes be derived from the same protein. Therefore, we envision the ability of TCRs to assume different conformations to recognize distinct antigenic surfaces to be a fundamental property of TCR recognition that increases the potential
diversity of the TCR repertoire.
6457