| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Microbiology and Immunology, Pennsylvania State University College of Medicine, Hershey, PA 17033; and
The Jackson Laboratory, Bar Harbor, ME 04609
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A molecular basis for MHC restriction has emerged from solving numerous
three-dimensional structures of class I molecules (reviewed in 3 .
The heterotrimeric class I molecule consists of the highly diverse
heavy chain, the invariant light chain ß2-microglobulin
(ß2m),3 and a
peptide. The 
1 and 2 domains of the heavy chain fold in such a
manner that they form an Ag-binding groove confined on the sides by two
-helices and at the bottom by two ß-sheets of four antiparallel
ß-strands each. Peptide Ags bind within this groove. The 3 domain
of the heavy chain and ß2m attain an Ig-like fold;
ß2m is noncovalently associated with the 1, 2,
and 3 domains (4, 5). Peptides bind the Ag-binding groove through
hydrogen-bonding interactions between the side chain atoms of the
conserved residues of pockets A and F of the heavy chain and the main
chain atoms of the amino-terminal amine and the carbonyl oxygen of the
carboxyl terminus of the peptide, respectively (6, 7, 8, 9, 10). Peptides that
bind class I molecules contain structurally conserved residues at the
carboxyl terminus (P
) and at an additional internal position. A
combination of these conserved residues within peptides constitutes the
class I binding motif (11, 12, 13). Thus H-2Kb class I
molecules bind peptides that predominantly contain phenylalanine (Phe)
or tyrosine (Tyr) at position (P) 5 and a hydrophobic aliphatic residue
(e.g., valine (Val), leucine (Leu), and isoleucine (Ile)) at P (11, 12, 14). Peptide binding studies using single amino acid variants of
antigenic peptides have revealed secondary anchors required for
effective binding (15, 16, 17). In the case of H-2Kb, residues
at P2 and P3 serve as secondary anchors (15, 18).
The crystal structure of class I molecules also revealed that the
majority of the amino acid residues that vary among class I allelic
products are those that contact the bound peptide by virtue of their
location in the Ag-binding groove. Thus the physico-chemical nature of
the Ag-binding groove dictates the anchor motif required for specific
peptide binding to a class I molecule (5, 19, 20). In the three crystal
structures of H-2Kb molecules complexed with different
octameric peptides, the dominant anchors, P5 Phe or Tyr, and the P
Leu are completely buried in the groove (8, 9, 21). Of the secondary
anchors, those at P2 are tucked into pocket B, and those at P3 are
almost completely buried within pocket D of the Ag-binding groove (8, 9, 21), consistent with the view that they can influence peptide
binding to H-2Kb. Thus, in octameric peptides displayed by
H-2Kb, residues at P2, P3, P5, and P are in intimate
contact with the class I molecule; the side chains at the remainder of
the peptide, i.e., at P1, P4, P6, and P7, are solvent exposed and
potentially accessible to the TCR (8, 9, 21).
Functional studies have suggested that the TCR interacts with a
composite structure of the class I-peptide complex in an orientation
diagonal to the -helices (22). In addition to confirming these
aspects of the TCR/class I-peptide interaction, the two recently solved
crystal structures of the TCR cocomplexed with its cognate ligand,
H-2Kb-self peptide in one and HLA-A2-HIV tax peptide in the
other, also revealed that the TCR interfaces Ag through its
complementarity determining regions (CDR) 1, 2, and 3 of the Tcr-
and Tcrß-chains (23, 24). Of these six CDRs, CDR3 makes the majority
of the contacts with the bound peptide (23), consistent with the fact
that CDR3, CDR3ß, and the peptide are the most variable parts of
the interacting surfaces. X-ray crystallographic analyses of TCR/class
I-peptide complexes have provided an in-depth view of how these
marcomolecules interact with each other. However, it provides a static
snap-shot of the most thermodynamically favored structure of the
interacting molecules. Therefore, further structure function analyses
are required to understand how the TCR interacts with its ligand.
Combinatorial libraries provide a powerful tool for probing the
interfaces of two interacting macromolecules (25, 26, 27, 28, 29, 30, 31). In this study,
combinatorial peptide libraries were used to gain insight into the role
of H-2Kb contact and TCR accessible residues in Ag
presentation and recognition. The results revealed that the display of
Ag to the TCR critically depended on the class I-contact residue at P3
of the peptide. In several instances the amino acid residue at P3 was
more critical than the dominant P anchor. Additionally, the TCR
requires only one or two TCR-contact residues of the peptide to achieve
specificity, while it can interact with multiple structurally
dissimilar amino acids of the remaining solvent-exposed residues.
Although the critical TCR contact residues seemed predisposed to the
carboxyl terminus of the peptide, the position of these residues was T
cell, therefore TCR, dependent. The requirement for only a few specific
TCR contact residues within the peptide explains why TCR/MHC-peptide
interactions are of low affinity and degenerate.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RMA cells (32) were used as targets in cell-mediated lympholysis. These cells were maintained in RPMI 1640 (Life Technologies, Gaithersburg, MD) supplemented with 5% heat-inactivated FCS, 100 U/ml penicillin, and 100 µM streptomycin. The H-2Kb-restricted, H-4b-specific CTL line M9 was isolated from C57BL/10 mouse, primed, and restimulated with H-4-congenic B10.129-H-4b (21 M) splenocytes. A cloned line was established and maintained with weekly stimulation using irradiated 21 M-derived spleen cells in the presence of 30 U/ml recombinant IL-2, as described (33).
The H-2Kb-restricted, gB-specific CTL clone 2D5 and the
RR1-specific CTL clone 1D11 were derived from splenocyte
populations of HSV-I immunized C57BL/6 mice (34, 35). Both the 2D5 and
1D11 CTL clones were maintained with 10% Rat T-Stim culture supplement
(Collaborative Biomedical Research Products, Bedford, MA) along
with 0.05 M -methyl-D-mannoside (Sigma, St. Louis,
MO) and stimulated with mytomycin C-treated, HSV-I-infected
B6/WT-3 cells (36), as described previously (34, 35).
Peptides
All combinatorial peptide libraries were produced by automated
solid phase F-moc synthesis (Chiron Technologies, Emeryville,
CA). The peptide libraries were synthesized with Phe at P5 in
all cases and either a single amino acid except Cys at defined
positions (solid circles in the Figures) or with equimolar mixtures of
all naturally occurring L-amino acids except for Cys for
randomized positions (x). The peptide preparations were controlled for
purity by random amino acids analysis, reversed-phase chromatography,
and electrospray-ionization mass spectrometry (Chiron Technologies).
Stock solutions of the peptide were prepared in DMSO at 
20 mg/ml.
All peptides were diluted in RPMI 1640 containing 5% FCS for the CTL
assay. Note that heat-inactivated FCS has low concentrations of
proteases, hence suggested for use in cell cultures where peptide
stability is important (37). For libraries 1 through 6, 5 µl of
peptide library stock were resuspended in 1.3 ml RPMI 1640 (76.9
µg/ml) whereas 5 µl of stock of libraries 7 through 11 were
resuspended in 1 ml RPMI 1640 (100 µg/ml) and used to sensitize
51Cr-labeled RMA cells. The number of peptides in each
library and the concentration of individual peptides in each library in
the assay are presented in Table I
.
|
Chromium release assays were performed according to standard
protocols. The E:T ratios for H-4b-specific, gB-specific,
and RR-1-specific CTL assays are described in the Figure legends. RMA
target cells (2 x 105 cells/ml) were preincubated for
2 h with an equal volume of peptide solution in RPMI 1640
supplemented with 5% FCS (see Table I
for final peptide concentration
in the assay). After 2 h at 37°C, an equal volume of effector
cells in RPMI 1640 supplemented with 5% FCS was added and incubated at
37°C for 5 to 6 h in V-bottom, 96-well plates. One half of the
supernatant in each well (100 µl) was harvested and monitored for
radioactivity (Beckman, Fullerton, CA). Specific lysis was
calculated as the percentage of the ratio of the difference between the
experimental and the spontaneous release to the difference
between the maximum and the spontaneous release.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A secondary anchor residue at P3 plays a key role in Ag presentation to class I-restricted CTL clones
The initial goal was to determine the role of P anchor residue
in H-4b Ag presentation. Thus, library 1 consisting of
octameric peptides with an invariant Phe at P5 (P5F), all naturally
occurring L-amino acids except cysteine (Cys) at P1, P2,
P3, P4, P6, and P7, and either Phe, Ile, Leu, methionine (Met), Tyr, or
Val at P (Fig. 1A), was
used to determine the P anchor. Cys was omitted to avoid
complications that could arise from covalent modification of the free
sulfydryl group of this residue. The resulting library consisted of six
sets of peptides distinguished by the amino acid residue at P and
contained 47,045,881 peptides per set. Peptides were presented to the
H-4b-specific CTL clone by RMA cells (described in
Materials and Methods); RMA cells express
H-2KbDb molecules of C57BL/6 origin but do not
express the BALB.B-derived H-4b minor H Ag (data not
shown). The H-4b-specific CTL recognizes H-2Kb
overexpressed by a BALB/c-derived plasmacytoma, Kb-high (39) (data not
shown). Surprisingly, none of the six peptide sets within this library
were recognized by the H-4b-specific CTL clone (Fig. 1A).
Peptides in library 1 may not have sensitized H-4b-specific
CTL targets for the following reasons. First, the peptides in the
library may have been insoluble, hence not presented to the CTL.
However, all six peptide sets in library 1 were presented equally well
to a C57BL/6-derived polyclonal population of CTL directed against at
least four distinct H-2Kb-restricted minor H Ags, including
H-4b, of BALB.B origin (data not shown). Thus, solubility
is less likely to be the reason why this CTL clone did not recognize
peptides of library 1. Second, concentration of individual peptides in
library 1 might have been too low to sensitize target cells for lysis
by the H-4b-specific CTL clone. However, the low
concentration of peptides in library 1 may not be the reason for the
lack of recognition. This conclusion is based on the fact that the
polyclonal CTLs raised in C57BL/6 mice against BALB.B splenocytes
recognized peptides in library 1 very efficiently (40% specific lysis;
data not shown). Moreover, peptide library 1 is recognized with a high
efficiency by a CTL clone directed against HSV-I ribonucleotide
reductase (RR-I)-derived epitope (described below; see Figure 5). Thus,
neither concentration nor solubility could account for the lack of
recognition of peptides in library 1 by the H-4b-specific
CTL clone.
|
anchor residue
but was not meant to replace the P anchor, which is at P8 of
Kb-binding peptides. That the Gly-extended libraries did
not replace the P anchor is substantiated by the crystal structure
of HLA-A2 complexed with a decamer peptide that has an additional Gly
residue at P10 of a naturally processed nonamer peptide (40). In this
structure, Gly at P10 extended out of the Ag-binding groove without
replacing the original P anchor at P9 (40). The naturally processed
peptides longer than nonamers bind HLA-A2 with high affinity (41) and
are stable at the physiologic temperature (40). Thus, we predicted that
the extension of a normally octameric peptide epitope(s) by a Gly at
the carboxyl terminus could provide a combinatorial library to test the
role of a hitherto unknown secondary anchor.
Library 2 is distinct from library 1 in that it consists of a mixture
of Phe, Ile, Leu, Met, Tyr, and Val at P8 and has one of the following
eleven amino acids, alanine (Ala), Phe, Gly, Ile, Leu, Met, asparagine
(Asn), proline (Pro), glutamine (Gln), Val, or Tyr, at P3. The residues
at P3 distinguish each peptide set in library 2 (Fig. 1B).
The choice of residues at P3 was based on the already known anchor
motif of H-2Kb-binding peptides (11, 14). The remaining
positions, P1, P2, P4, P6, and P7, of the peptide contained all
naturally occurring L-amino acids except Cys (Fig. 1B). One peptide set, distinguished by the presence of Ile
at P3 (P3I), among the eleven sets of peptides in library 2 was
recognized by the H-2Kb-restricted,
H-4b-specific CTL clone (Fig. 1B). To determine
whether the Gly extension at the carboxyl terminus contributed to the
CTL activity of the peptides in library 2, library 3 was synthesized.
It is essentially similar to library 1 but extended by a Gly residue at
the carboxyl terminus (Fig. 1C). Although this library was
presented to the polyclonal CTL against minor H Ags described above
(data not shown), none of the six sets of peptides in library 3 were
recognized by the H-4b-specific CTL clone (Fig. 1C). Hence, the Gly extension does not contribute to the
recognition of peptides in library 2 by H-4b-specific CTL
clone. Together the data suggest that the amino acid residue at P3 can
critically affect the presentation of peptide Ags by class I molecules
to specific CTL.
Role of secondary anchor residue at P2 in Ag presentation to class I-restricted CTL clones
P anchor of H-4b-derived CTL epitope.
To determine the amino acid residue(s) at P2 in the
H-4b-derived epitope, the P anchor of H-4b
Ag had to be defined. Therefore, library 4 of octameric peptides was
constructed. It consists of five sets of peptides related to each other
by virtue of containing P3I and P5F. The peptide sets are distinct from
each other because each set of 2,476,099 peptides contains Phe, Ile,
Leu, Met, or Val at P (Fig. 2A). P1, P2, P4, P6, and P7
are degenerate, containing all the natural L-amino acids
but Cys (Fig. 2A). Once an invariant amino acid residue was
fixed at P3 (P3I) in the peptide library, the residue at P was
easily identified. Thus the H-4b-derived epitope consists
of either Val, Leu, or Ile at P (Fig. 2A). Additionally,
two similar libraries, P3G-containing library 5 (Fig. 2B)
and P3F-containing library 6 (Fig. 2C), consisting of four
and five octameric peptide sets based on the different amino acid
residues at P, respectively, were not recognized by the
H-4b-specific CTL clone (Fig. 2B and 2C). These
data further emphasize the stringent requirement of P3I in the
structure of the H-2Kb-restricted H-4b epitope.
|
) at P2. Trp was omitted because none of
the H-2Kb-restricted peptides reported to date contain this
residue at P2 (13). Additionally, peptides in library 7 consisted of
invariant P3I, P5F, and PL and all natural L-amino acids
except Cys at P1, P4, P6, and P7 (Fig. 3). Note that, although Ile,
Leu, and Val could be used as the P anchor, PL was chosen because
Leu is most frequently used as the carboxyl terminus amino acid residue
in peptide epitopes presented by H-2Kb molecules (11, 13).
The results revealed that the H-4b-derived epitope
accommodates only Gly at P2, without affecting CTL recognition (Fig. 3). Thus, the secondary anchor at P2 of H-4b-specific CTL
epitope is invariant.
|
Both functional and crystal structure studies of three
H-2Kb-restricted antigenic peptides have revealed that
amino acid residues at P1, P4, P6, and P7 are solvent exposed and
oriented in space in a manner accessible by specific TCR (8, 9, 21).
The amino acid structure requirements of the four putative
TCR-accessible residues for H-4b Ag recognition were
determined using libraries 8, 9, 10, and 11. These libraries were
similar to library 7 in that the 123,462 peptides in each library
contained P3I, P5F, and PL. In addition, three of the four putative
TCR contact positions and P2 contained a mixture of all the natural
L-amino acids except Cys while the fourth position
contained one of the 20 natural L-amino acids except Cys
(Fig. 4). Thus library 8 has one of the
20 natural L-amino acids except Cys at P1 and a mixture of
all of the natural L-amino acids except Cys at P2, P4, P6,
and P7, and so on (see Figure 4).
|
). It recognizes peptides containing multiple structurally
distinct amino acid residues at P1 (Glu, Gly, and Gln), P4 (Ile, Tyr,
and Val) and P7 (arginine (Arg) and histidine (His); Fig. 4, A–C). However, this CTL clone recognizes Ag only
when it contained an invariant Ile at P6 (Fig. 4D). Thus,
although degenerate, the specific TCR interfaces Ag only when presented
with a common structurally conserved amino acid residue at one position
in the CTL epitope that is solvent exposed and accessible to the TCR
(Fig. 4E). How general are the rules?
The generality of the rules regarding H-2Kb-restricted
Ag presentation and recognition defined using the
H-4b-specific CTL clone were further tested using two CTL
clones whose epitopes are known. One CTL clone is specific for an
immunodominant herpes-simplex virus (HSV)-I’s glycoprotein B
(gB)-derived epitope, SSIEFARL (35). The second CTL clone is
directed against an immunorecessive HSV-I’s ribonucleotide reductase-1
(RR-1)-derived epitope, QTFDFGRL (34). These two CTL clones were chosen
because both their epitopes contain P5F and PL, while one contains
P3I (gB) and the other contains P3F (RR-I) (34, 35). The gB- and
RR-1-specific CTL clones do not recognize RMA cells unless pulsed with
their respective epitopes (data not shown); they are sensitive to
picomolar amounts of the peptide (34, 35).
The P3 rule.
The stringent requirement of an invariant residue at P3 observed with
H-4b-specific CTL clone was tested using the RR-1-specific
CTL clone. Unlike H-4b-specific CTL clone, the
RR-1-specific CTL clone recognized library 1 (Fig. 5A) but did not recognize
libraries 2 and 3 (data not shown). In fact, P3F of the native epitope
could be substituted by Gly or Ile without affecting recognition by
RR-1-specific CTL clone (Fig. 5, B, C, and
D). Although RR-1-specific CTL clone recognized peptide
library 6 containing P3I, it did not significantly recognize any of the
peptides represented in libraries 7 and 8; the peptides in these two
libraries contained invariant P3I, P5F, and PL (Fig. 5, E
and F). Note that the background in the assay presented in
Figure 5 was high; in a repeat of the experiment using libraries 7 and
8, none of the peptide sets in these two libraries were recognized
(data not shown). Additionally, multiple structurally distinct
residues, including those not represented in the native epitope, were
recognized by the RR-1-specific CTL clone when used to probe libraries
9, 10, and 11, which also contained P3I, P5F, and PL (Fig. 5, G, H, and J). A summary of the
RR-1-specific CTL recognition pattern is represented in Figure 5K. Together with the H-4b-specific CTL
recognition pattern, these data suggest that the residue at P3
critically affects Ag presentation by H-2Kb to specific
CTL.
Unlike the H-4b- and RR-1-specific CTL clones, none of the
peptides within libraries 1 through 6 were recognized by the
gB-specific CTL clone (data not shown); note that these libraries have
fixed anchors only at P5 and P or at P3 and P5. However, peptides
within libraries 7 through 11, containing P3I, P5F, and PL as
invariant residues, sensitized targets for recognition by this CTL
clone (described in detail below; Figure 6). This suggests that the three anchors
P3I, P5F, and PL together are essential in presenting the gB-derived
epitope to specific CTL.
|
A); note that the natural gB-derived
epitope contains Ser at P2. These data are consistent with the finding
that OVA-derived epitope, SIINFEKL, can accommodate Ala, Gly, Leu, Ser,
Thr, and Val at P2 without affecting recognition by certain
H-2Kb-restricted, OVA-specific CTL clones (18). Thus
multiple structurally distinct amino acid residues can be accommodated
at P2 of some H-2Kb-restricted CTL epitopes.
The TCR contact rule.
This rule was tested using the H-2Kb-restricted,
gB-specific CTL clone. Akin to H-4b-specific CTL clone,
gB-specific CTL clone also recognized peptides containing multiple
structurally distinct amino acid residues at P1 (Phe, Gln, Ser, and
Trp, as well as Asp and Thr to a lesser extent; Fig. 6B) and
P6 (Ala, Phe, and Tyr, as well as Trp to a lesser degree; Fig. 6D). At the remaining positions, the gB-specific CTL clone
recognized only the residues represented in the native epitope, i.e.,
Glu at P4 and Arg at P7 (Fig. 6, C and E). These
data are summarized in Figure 6F. Thus, although degenerate,
the specific TCR interfaces Ag only when presented with common
structurally conserved amino acid residues at one or two positions
in the CTL epitope among the four putative solvent-exposed residues
that may be accessible to the TCR.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1.7 x 1010 individual
peptide combinations for CTL recognition. Thus the combinatorial
peptide library approach as reported here and by others provides a
solution to massive screening for CTL recognition. The combinatorial library approach is not without limitations. First, it is difficult to ascertain the solubility of individual peptides in a complex library. The solubility problem was addressed in this study using a polyclonal population of CTL against BALB.B-derived multiple minor H Ags, which also includes the anti-H-4b reactivities, to probe the libraries. The results revealed that a reasonable proportion of the peptides in the libraries tested were in solution. Second, competition for binding class I could exclude peptide(s) in a library from being presented. However, fewer than 100 class I-peptide complexes are typically necessary for CTL recognition (45, 46); hence, this is unlikely to be a problem. Third, each peptide library would be expected to contain an admixture of agonistic and antagonistic peptides for a specific TCR (27). The recognition of the targets sensitized by peptides in a complex library by CTL, then, represents the effect of those agonistic peptides not inhibited by the antagonistic peptides in the library. Whether a set of antagonist peptides in library 1 inhibited the recognition of agonist peptides in this library by the H-4b-specific CTL clone could not be tested because the H-4b-derived epitope hitherto remains elusive. Notwithstanding, the combinatorial peptide library approach can provide useful information regarding TCR/class I-peptide interactions.
Thus, synthetic combinatorial peptide libraries were used as an
approach to define the rules for class I-restricted peptide Ag
presentation and recognition. To avoid bias from the knowledge of a CTL
epitope, the physico-chemical features of the unknown
H-2Kb-restricted, H-4b Ag-specific CTL epitope
was first determined using the combinatorial libraries. As expected,
several mimotopes were identified. The physico-chemical features of the
H-4b-derived mimotopes were compared with those of
similarly derived mimotopes of two other H-2Kb-restricted,
viral Ag (HSV-I gB and RR-1)-specific CTL clones whose epitopes are
already known. This allowed the definition of two important structural
features for class I-restricted peptide Ag presentation and
recognition: 1) The display of Ag to H-2Kb-restricted TCR
critically depended on the class I-contact residue at P3 of the
peptide. In several instances the amino acid residue at P3 was more
critical than the dominant P anchor. 2) The TCR requires only one or
two TCR-contact residues of the peptide to achieve specificity, while
it can interact with multiple structurally dissimilar amino acids at
the remaining solvent-exposed positions.
Role of secondary anchor residues in Ag presentation
The role of dominant anchor residues, P5 and P in the case of
H-2Kb-binding peptides, is thoroughly appreciated in
numerous previous studies. Although the secondary anchor residues play
an important role in peptide binding to class I molecules, their role
in presentation of antigenic peptides to CTL is less clearly
understood. A recent study demonstrated that the recognition of OVA
epitope SIINFEKL by H-2Kb-restricted, OVA-specific CTL
clones was altered by the structural features of the residues at P2.
Nonetheless, although clone specific, these CTL clones tolerated
structurally distinct amino acids (Ala, Asn, Gly, Ser, and Thr) at P2
in the SIINFEKL epitope (18). This is consistent with the finding
presented here, in that the gB-specific CTL clone tolerates
structurally distinct amino acid residues at P2 in its epitope. In
contrast to the degeneracy at P2 in gB-specific CTL epitope,
recognition by H-4b-specific CTL requires an invariant Gly
residue at P2. On the other hand, appropriate presentation of
H-4b and RR-1 Ags by H-2Kb critically depended
on the nature of the residue at P3. Thus, residues at P3 serve as an
important secondary anchor at least in appropriate presentation of Ag
to CD8+ T cells studied here. The side chain of the residue
at P3 is oriented toward the 2-helix and almost completely buried
into pocket D in the three crystal structures of H-2Kb
individually complexed with different peptides (8, 9, 21). Thus,
residues at P3 probably assume the role of an important secondary
anchor.
A thorough x-ray crystallographic analysis of HLA-A2 individually complexed with five different peptides revealed that the secondary anchors can influence the orientation of the solvent-exposed residues without altering the conformation of the dominant anchors. Importantly, none of the peptides induced any significant variation in the HLA-A2 structure. Because the solvent-exposed peptide residues are accessible to the TCR, it was concluded that interactions with such residues would determine the antigenicity of the class I-peptide complex (47). Thus the results presented herein provide functional evidence for the role of secondary anchor residues, such as those at P3, in altering the antigenicity of the class I-peptide complex.
A model for TCR/class I-peptide interactions
The two recently solved crystal structures of TCR/class I-peptide cocomplexes have provided detailed information on how the receptor interfaces its ligand (23, 24). Based on these structures, we would predict that the CDR3ß loop interfaces the carboxyl half of the peptide Ags derived from H-4b, gB, and RR-1. In this regard, it is noteworthy that the majority of the TCR expressed by over 15 distinct CTL clones against H-4b (48) and gB (49) contain an invariant acidic residue in their CDR3ß loop. Being of opposite charge, the acidic residue in CDR3ß can make a productive interaction with the basic Arg (H-4b and gB) or His (H-4b) residues present in the respective epitopes. This interaction may be important but does not impart the recognition specificity between the receptor and the ligand because the two CTL clones do not have overlapping epitopes or mimotopes. Thus, P6I and P4E of the H-4b- and gB-specific CTL epitopes, respectively, may play a key role in imparting specificity of the recognition process. Thus, very few atoms of the epitopes are essential for the specificity in Ag recognition by the TCR.
Although much has been learned regarding how the TCR interfaces its Ag,
the basis for the degenerate nature of the TCR/class I-peptide
interactions remains to be established. Several studies have
established that the interaction of the TCR with its ligand is of low
affinity (50, 51, 52). These affinity values are similar to those for the
interaction between IgM and its Ag and contrast that of IgG, which has
a much higher affinity for its Ag (53, 54, 55). Consequently, the TCR/class
I-peptide interactions occur with a slow on and fast off rates (50, 51). Calculations of the Gibbs free energy change accompanying the
formation of Ab and Ag complexes revealed that the higher the affinity
of the interaction, the greater the binding energy. Additionally, both
the affinity and the binding energy are directly proportional to the
number of amino acid residues that are critical in the Ab-Ag
interactions (56). Thus, by analogy, the binding energy accompanying
the ternary complex formation would be minimal, based on the physical
parameters of TCR/class I-peptide interactions. The low binding energy
of this reaction would suggest that the specificity of the TCR/class
I-peptide binding is imparted by interactions between few atoms of the
interacting components with the remaining large (600
Å2) interface area probably participating in
physico-chemical complementarity.
In support of the above model is the finding that the TCR of the H-4b- and the gB-specific CTL clones require only one of the four solvent-exposed residues in their cognate epitopes for effective Ag recognition. Additionally, this model would also assume a strict spatial positioning of the critical TCR contact residue(s) in relation to the remaining interface area on the class I-peptide complex for effective TCR/class I-peptide interactions. Thus, alterations in the strict spatial order of the critical TCR contact residues by differences in the secondary anchors, as seen with the H-4b- and RR-1-specific TCR, can affect TCR/class I-peptide interaction patterns. Finally, the above model for TCR/class I-peptide interaction also explains how a single TCR interfaces multiple ligands.
Our finding that the TCR contacts only one or two of the four solvent-exposed residues of the peptide to achieve specificity is supported by the recently described crystal structure of the murine 2C TCR complexed with H-2Kb bound to self-peptide dEV8 (52). The dEV8 peptide is buried deep in the H-2Kb groove, hence contributing very little to the overall TCR/MHC-peptide interactions. Moreover, the interface between the 2C TCR and the H-2Kb-dEV8 complex has poor three-dimensional complementarity and has large empty spaces, making very few contacts with the peptide possible (52). Similarly, the A6 TCR/HLA-A2-HIV tax cocomplex also revealed poor structural complementarity at the interface between the receptor and its ligand (23). Thus, the poor complementarity observed in the above crystal structures explains why the TCRs studied here tolerate multiple structurally dissimilar amino acids at most of the TCR contact residues of the peptides presented by H-2Kb.
In conclusion, akin to the class I-peptide interactions whose specificity is determined by the structure of the residues at select positions in the peptide (anchor residues), the TCR/class I-peptide interaction also requires only an amino acid or two among at least four available TCR contact residues within the peptide for specificity. In this manner a single TCR can interface multiple ligands.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Sebastian Joyce, Department of Microbiology and Immunology, Pennsylvania State University College of Medicine, Milton S. Hershey Medical Center, Hershey, PA 17033. E-mail address:
3 Abbreviations used in this paper: ß2m, ß2-microglobulin; CDR, complementarity determining region; gB, glycoprotein B; HSV-I, herpes simplex virus type I; RR-1, ribonucleotide reductase-1.
Received for publication March 20, 1998. Accepted for publication June 22, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Yoshimura, R. Yadav, G. J. Christianson, W. U. Ajayi, D. C. Roopenian, and S. Joyce Duration of Alloantigen Presentation and Avidity of T Cell Antigen Recognition Correlate with Immunodominance of CTL Response to Minor Histocompatibility Antigens J. Immunol., June 1, 2004; 172(11): 6666 - 6674. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. W. H. G. Kessels, K. E. de Visser, F. H. Tirion, M. Coccoris, A. M. Kruisbeek, and T. N. M. Schumacher The Impact of Self-Tolerance on the Polyclonal CD8+ T Cell Repertoire J. Immunol., February 15, 2004; 172(4): 2324 - 2331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Yadav, Y. Yoshimura, A. Boesteanu, G. J. Christianson, W. U. Ajayi, R. Shashidharamurthy, A. K. Stanic, D. C. Roopenian, and S. Joyce The H4b Minor Histocompatibility Antigen Is Caused by a Combination of Genetically Determined and Posttranslational Modifications J. Immunol., May 15, 2003; 170(10): 5133 - 5142. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Brehm, T. G. Markees, K. A. Daniels, D. L. Greiner, A. A. Rossini, and R. M. Welsh Direct Visualization of Cross-Reactive Effector and Memory Allo-Specific CD8 T Cells Generated in Response to Viral Infections J. Immunol., April 15, 2003; 170(8): 4077 - 4086. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. E. de Visser, T. A. Cordaro, H. W. H. G. Kessels, F. H. Tirion, T. N. M. Schumacher, and A. M. Kruisbeek Low-Avidity Self-Specific T Cells Display a Pronounced Expansion Defect That Can Be Overcome by Altered Peptide Ligands J. Immunol., October 1, 2001; 167(7): 3818 - 3828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Joshi, P. R. Suresh, and V. S. Chauhan Flexibility in MHC and TCR Recognition: Degenerate Specificity at the T Cell Level in the Recognition of Promiscuous Th Epitopes Exhibiting No Primary Sequence Homology J. Immunol., June 1, 2001; 166(11): 6693 - 6703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. C. Godthelp, M. J. D. van Tol, J. M. Vossen, and P. J. van den Elsen Longitudinal analysis of T cells responding to tetanus toxoid in healthy subjects as well as in pediatric patients after bone marrow transplantation: the identification of identical TCR-CDR3 regions in time suggests long-term stability of at least part of the antigen-specific TCR repertoire Int. Immunol., April 1, 2001; 13(4): 507 - 518. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Regner, M. Lobigs, R. V. Blanden, P. Milburn, and A. Mullbacher Antiviral Cytotoxic T Cells Cross-Reactively Recognize Disparate Peptide Determinants from Related Viruses but Ignore More Similar Self- and Foreign Determinants J. Immunol., March 15, 2001; 166(6): 3820 - 3828. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Sandberg, L. Franksson, J. Sundback, J. Michaelsson, M. Petersson, A. Achour, R. P. A. Wallin, N. E. Sherman, T. Bergman, H. Jornvall, et al. T Cell Tolerance Based on Avidity Thresholds Rather Than Complete Deletion Allows Maintenance of Maximal Repertoire Diversity J. Immunol., July 1, 2000; 165(1): 25 - 33. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
