| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Institute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, GA 30912
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The interaction of the individual TCR with self MHC/peptide complexes determines the fate of the immature T cell resulting in positive or negative selection and emphasizes the importance of these interactions to the final repertoire of TCRs. Two properties of the preselected TCRs, their intrinsic capacity to interact with MHC molecules and their specificity to self peptides bound to MHC molecules, account for the success or failure of the individual thymocyte in thymic selection. It is estimated that between 5 and 20% of preselected TCRs have an inherent capacity to interact with a given MHC, based on different experimental protocols (9, 10, 11). It is expected, therefore, that the population of self-derived peptides bound to the thymic MHC may influence the efficiency and specificity of thymic selection. There are many experimental results proving that the self-derived peptides are specifically recognized during negative selection of T cells (reviewed in 12 . In contrast, the extent to which the recognition of self peptides bound to thymic MHC during positive selection is peptide specific or degenerate remains controversial (13).
For example, experimental protocols that directly investigated the peptide contribution to positive selection of class I MHC-restricted CD8+ T cells employed fetal thymus organ cultures (FTOC)3 from either ß2m- or TAP-deficient mice that normally do not express class I MHC and have severely impaired positive selection of CD8+ T cells (14, 15). However, the expression of class I MHC could be restored by supplementation with a soluble peptide(s) capable of binding to the appropriate class I MHC (FTOC from ß2m- mice also required exogenous ß2m). Using this technique, the in vitro selection of CD8+ T cells by class I MHC molecules loaded with the tested peptide(s) could be followed. These models demonstrated that class I MHC molecules loaded with the mixture of exogenously added peptides are more efficient in positively selecting CD8+ T cells than the same MHC molecules loaded with any single peptide tested (14, 15). Moreover, two different transgenic TCRs were found to be positively selected by class I MHC molecules loaded with synthetic peptides with primary sequences similar to the antigenic peptides. On the other hand, same MHC molecules loaded with peptides unrelated to antigenic peptide were unable to select CD8+ TCR transgenic thymocytes (16, 17, 18). From these results, the authors concluded that positive selection is highly specific for peptide ligand and therefore very restricted (16, 19). Such an interpretation was questioned by experiments describing efficient positive selection of CD8+ T cells in TAP-deficient/ß2m transgenic mice, when a diverse set of CD8+ T cells was selected by class I MHC molecules bound to a tightly limited set of peptides (20). Furthermore, another transgenic TCR expressed on CD8+ T cells was positively selected in FTOC on synthetic peptides structurally different from the antigenic peptide (21). These last results argued for promiscuous rather then peptide-specific recognition during positive selection of CD8+ T cells.
Independently, the specificity of positive selection of class II MHC-restricted CD4+ T cells was primarily studied by in vivo techniques. Two separate models, one based on mice that express single class II MHC molecules (Ab) bound with peptide (Ep) via the flexible linker (AbEpIi- mice) and the second based on mice that lack DM molecules that dislocate CLIP (class II-associated invariant chain peptide) from class II MHC (DM-deficient mice (DM-)), showed that about 15–50% of the normal number of CD4+ T cells are selected in these animals, respectively (22, 23, 24, 25). The repertoires of TCRs on CD4+ T cells selected by single (AbEpIi-) or dominant (DM-) class II MHC/peptide mice used various TCR Vß segments, which suggested that TCRs may encode various Ag specificities. However, it was also revealed that approximately two-thirds of CD4+ T cells from AbEpIi- mice respond to wild-type Ab/peptide complexes in vitro, which may lead to their negative selection in normal mice in vivo (22). Therefore, <5% of the normal number of CD4+ T cells can be positively selected on a single class II MHC/peptide complex and potentially can be included in the normal repertoire. Ag specificities encoded by TCRs positively selected by one, but negatively selected by multiple, class II MHC/peptide complexes remain to be determined.
The finding that one class II MHC/peptide complex can positively select various TCRs was further confirmed by experiments where an intrathymic injection of a recombinant adenovirus led to the expression of an Ek molecule bound with neopeptide exclusively on the thymic epithelial cells (26). The repertoire of TCRs on CD4+ T cells selected by Ek/neopeptide was heterogeneous. Undoubtedly, then, some class II MHC/peptide complexes can select a number of different TCRs, but not every TCR. For example, several transgenic TCRs found to be selected on wild-type Ab/self peptide complexes were not positively selected in either DM- mice or AbEpIi- mice (27, 28). Thus, although the repertoires of CD4+ T cells selected in thymi expressing single or dominant class II MHC/peptide complexes are more diverse than previously predicted, these repertoires are limited in volume and probably in the number of encoded Ag specificities.
All tested models investigating thymic selection of either
CD4+ or CD8+ T cells agreed that the wild-type
repertoire of TCRs requires a diverse set of self derived peptides to
be involved in thymic selection. It is still questionable how many
different Ag specificities are encoded by the set of TCRs selected by
the defined single MHC/peptide complex and are potentially present in
the normal repertoire. Recent studies have shown that, in contrast to
previous expectations, the selecting and antigenic peptides recognized
by the same TCR do not have to be related in primary amino acid
sequence (26, 29, 30). Moreover, analysis of the repertoire of
CD4+ T cells selected in DM- radiation
chimeras reconstituted with wild-type bone marrow suggested that these
CD4+ T cells can respond to allogenic MHC molecules and
other proteins (28). The interpretation of this last result, however,
was complicated by a finding indicating that the residual self peptides
present in DM- mice may significantly contribute
to positive selection of CD4+ T cell in these mice (27).
Hence, to determine the physiological outcome of positive selection on
a single class II MHC peptide complex, we have studied the Ag
specificities of CD4+ T cells positively selected in vivo
by a single AbEp complex and negatively selected by
wild-type Ab/peptide complexes. We found that negative
selection on the normal spectrum of self class II MHC/peptide complexes
deletes more than half the CD4+ T cells positively selected
by AbEp. The remaining repertoire of CD4+ T
cells is semidiverse and is not significantly biased in
CD4+ T cells specific to antigenic peptides with primary
amino acid sequences close to Ep. Analysis of the DNA sequences of the
TCRs selected on the AbEp complex and specific for
different antigenic peptides revealed that this repertoire consists of
a number of 
- and ß-chains, some of which are similar. This result
indicates that in the normal repertoire, positive selection on a single
class II MHC/peptide complex contributes a limited set of TCRs with a
broader than expected repertoire of antigen specificities.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice expressing the single AbEp complex were produced at the National Jewish Center (Denver, CO) as described previously (22). These mice and mice deficient for class II MHC or invariant chain as well as AbEpIi-ß2m- mice were further bred in the animal care facility at the Medical College of Georgia (Augusta, GA). Mice expressing single AbPCC43–58 complex were bred at the transgenic facility of the Medical College of Georgia (Augusta, GA). C57BL/10 (Abwt) mice were purchased from The Jackson Laboratory (Bar Harbor, ME).
Chimeric mice were generated by irradiation of 6- to 12-wk-old animals (950 rad) followed by immediate i.v. reconstitution with 5 x 106 fetal liver cells from donor fetuses (day 15 gestational age). Chimeric mice were used for experiments no earlier than 8 wk after reconstitution.
Cell staining
Cells were stained for CD4, CD8, and different V and Vß
segments of TCR as previously described (31). The fluorescein- and
phycoerythrin-labeled Abs were prepared in our laboratory or purchased
from PharMingen (San Diego, CA). Cells were suspended in staining
buffer (BSS, 0.1% sodium azide, and 2% FBS) and were incubated for 30
min at 4°C with the Abs of interest in the presence of 10% normal
mouse serum and 10% culture supernatant of anti-Fc receptor Ab
2.4G2 (32) to block nonspecific binding. Cells were then washed three
times with staining buffer and analyzed using a FACSCalibur instrument
(Becton Dickinson, San Jose, CA).
Quantitative analysis of naive T cells
Cell suspensions were prepared from the spleen and the pool of inguinal, axillary, and mesenteric lymph nodes. The number of naive T cells was evaluated as previously described (33). In brief, the number of CD4+CD44lowCD45RBhigh T cells in each organ was calculated from the number of cells recovered using immunofluorescence analysis. The total number was obtained by adding the number of CD4+CD44lowCD45RBhigh T cells from spleen and twice that from lymph nodes.
Production of T cell hybridomas
The Abwt
AbEpIi- chimeras
were primed s.c. at the base of the tail with different synthetic
peptides in CFA; 7 days later the draining lymph nodes were harvested.
Lymph node cells were cultured for 3 days in culture medium in the
presence of the antigenic peptide (50 µg/ml). On day 3 dead cells
were removed by gradient centrifugation on Lymphocyte Separation Media
(Cellgro, Herndon, VA). The recovered live cells were expanded in the
presence of IL-2 for 3 days and converted into T cell hybridomas as
previously described (34). Hybridomas were analyzed for the expression
of CD4 and TCR. Double-positive hybridomas were screened for
peptide-specific response using an HT-2 assay (35). Briefly, 1 x
105 hybridoma cells were incubated with 5 x
105 splenocytes from Abwt mice with or without
peptide. After 24 h, the amount of secreted IL-2 was measured
using the detector cell line HT-2. The proliferation of HT-2 cells in
response to IL-2 was measured using
3-(4,5-dimethythiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT;
Sigma, St. Louis, MO) assay (36).
Analysis of V and Vß segments used by T cell hybridomas
The repertoire of Vß and V segments was established by flow
cytometric analysis of the hybridomas after staining with currently
available anti-mouse V and Vß reagents. RNA was isolated from
hybridomas using the Ultraspec RNA isolation kit (Biotecx, Houston, TX)
and was converted to cDNA using the RT system (Promega, Madison, WI).
TCR - and ß-chains were amplified from the cDNA by PCR using
constant region and V/Vß-specific primers (37). PCR products were
gel purified and sequenced by automated fluorescent sequencing on an
ABI Prism 377 DNA sequencer (PE Applied Biosystems, Foster City, CA).
Proliferation assays
The response of CD4+ T cells selected in
AbwtAbEpIi- chimeras was
analyzed by proliferation assay. Tested peptides were injected into
individual AbwtAbEpIi- chimeras
or wild-type mice in the base of the tail in the presence of CFA
(Sigma). On day 7 the lymph nodes from these mice were harvested, and
non-CD4+ T cells were eliminated by complement-mediated
lysis as follows. RBC were removed by incubation with buffered ammonium
chloride solution, and cells were resuspended in BSS to a final
concentration of 5 x 106 cells/ml and incubated for
30 min at 4°C in the presence of mAbs: YTS 169.4 (anti-CD8) and
25-9-3S (anti-I-Ab) at a concentration of 1–10
µg/ml. Cells were washed twice with BSS, resuspended to a final
concentration of 107 cells/ml and incubated for 45 min at
37°C with complement. After two washings, cell viability was
estimated by trypan blue exclusion. The purity of the CD4+
population was confirmed by FACS analysis.
Isolated CD4+ T cells were used in proliferation assays. CD4+ T cells were cultured for 6 days with irradiated splenocytes (3000 rad) in the presence of the specific antigenic or irrelevant peptides. On days 3 and 5, 20 IU of IL-2 were added to each well. The proliferative response was measured using the MTT assay as previously described (36).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In a previous study we analyzed the repertoire of CD4+ T cells selected in thymus expressing single class II MHC covalently bound with single peptide, Ep52–68 (30). Our investigation was complicated by a high incidence of reactivity to wild-type Abwt/peptide complexes in CD4+ T cells selected in AbEpIi- mice. We interpreted this property of CD4+ T cells selected in AbEpIi- mice to indicate a lack of tolerance to a diverse set of self-derived peptides presented by wild-type Ab molecules. Alternatively, it was suggested that an imprint of positive selection by a single AbEp complex makes this repertoire of CD4+ T cells so restricted that the reported reactivity to Abwt APCs reflects the recognition of one or a few, rather than many, Ab/self peptide complex(es) (38). It was also possible that the Ab molecule itself folds differently when bound with any peptide other than Ep, and this can activate a large number of CD4+ T cells selected in AbEpIi- mice.
To determine whether a diverse set of peptides bound by wild-type
Ab molecules produced the phenomenon described above,
we tested the proliferative responses of CD4+ T cells
selected in AbEpIi- mice to
Ab occupied with a different range of peptides. Purified
CD4+ T cells from
AbEpIi- mice were incubated with
irradiated spleen APCs expressing similar levels of Ab
molecules bound by many peptides (from AbEpIi+
mice), some peptides (from AbwtIi- mice), or
one peptide (AbPCC43–58Ii- mice) distinct from
Ep. After 5 days, the proliferation responses to these different APCs
displaying Ab/peptide complexes with various degrees of
peptide occupancy were estimated using the MTT incorporation assay. A
representative experiment (one of three) is shown in Fig. 1
. We found that APCs that express
Ab bound to many self peptides induced the strongest
proliferative response. The APCs derived from
AbwtIi- mice were weaker stimulators of
AbEp-selected CD4+ T cells, and virtually no
proliferative response to APCs from
AbPCCIi- mice was recorded. This
result indicated that the range of peptide diversity correlates with
the magnitude of the CD4+ T cell proliferative response.
However, whether few or many self-derived peptides are recognized by
CD4+ T cells selected by the AbEp complex
remained to be determined, since it is unknown to what extent
presentation of self peptides recognized by these CD4+ T
cells depends on the presence of invariant chain. To reconcile this
controversial issue it was requisite to determine the Ag specificities
of CD4+ T cells selected in
AbwtAbEpIi- radiation
chimeras, where CD4+ T cells were positively selected on
one and negatively selected on many Ab/self peptide
complexes (see next section).
|
, approximately 3% of thymocytes and
12% of peripheral T cells were CD4+ single positive in
AbEpIi-AbEpIi-
chimeras. In contrast, the number of CD4+ T cells was
severely reduced in
AbwtAbEpIi- chimeras.
In both thymus and lymph nodes, the reduction of CD4+ T
cells was approximately 50%, indicating ongoing negative selection on
wild-type Ab/self peptide complexes. This number is similar
to the number of Ab/self peptide-reactive cells in
AbEpIi- mice estimated in vitro as
well as the range of deletion observed in
AbwtDM- chimeras (28, 39).
Despite a prominent negative selection, around 3–5% of
CD4+ T cells found in wild-type mice were present in the
periphery in
AbwtAbEpIi- chimeras.
We assumed that this population includes CD4+ T cells
positively selected on the AbEp complex and therefore
focused on the complexity and Ag specificities of its TCRs.
|
AbEpIi- chimeras.
It was shown that the level of class II MHC in
AbEpIi- mice is 5–10 times lower
than that in wild-type mice due to the low expression of the transgenic
AbßEp construct (22). To investigate that issue, mice
expressing the single AbEp complex were reconstituted with
fetal liver from mice expressing the same AbßEp transgene
in the presence of an invariant chain
(AbEpIi+). The AbEpIi+
mice express virtually the same amount of Ab molecules on
the cell surface as AbEpIi- mice,
but in the presence of invariant chain the covalent Ep is extensively
cleaved and replaced with endogenous peptides (22). As also shown in
Fig. 2
, AbEpIi+AbEpIi-
radiation chimeras appeared to have only slightly more CD4+
T cells in the thymus and periphery than
AbEpIi- chimeras reconstituted with
wild-type fetal liver. This result suggested that it is primarily the
peptide content of class II MHC molecules rather than the difference in
MHC expression per se that is responsible for the deletion of
CD4+ T cells in
AbwtAbEpIi- chimeras.
Furthermore, CD4+ T cells from
AbEpIi+AbEpIi-
chimeras were tolerant to Abwt, as tested by the in vitro
response of the panel of T cell hybridomas (data not shown). This last
result suggested that the low level of transgenic Ab
molecules does not significantly hamper negative selection and that the
lack of tolerance to self peptides is what makes
AbEp-selected CD4+ thymocytes susceptible to
deletion by wild-type Ab/peptide complexes.
CD4+ T cells positively selected on AbEp
and tolerant to Abwt have a naive phenotype and bear a
heterogeneous set of ß TCRs
We next analyzed the surface phenotype of CD4+ T cells
selected in
AbwtAbEpIi- chimeras.
The purpose of this investigation was to examine whether exposure of
AbEp-selected CD4+ T cells to many self class
II MHC/peptide complexes alters the expression of some surface
molecules or reshapes their repertoire of ß TCRs. In particular,
we have analyzed the expression of T cell activation markers such as
CD44, CD69, and CD45RB, inspired by a recent report that the majority
of peripheral CD4+ T cells from
AbwtDM- chimeras have an
activated phenotype. Moreover, these CD4+ T cells were
found to be incapable of responding to in vitro stimulation with mAb
against TCR (28). Accordingly, we found that a number of
CD4+ T cells from
AbEpIi- mice and
AbwtAbEpIi- chimeras
also express the phenotypic profile of activated T cells (Fig. 3). We have sorted
CD4+CD45RBlow and
CD4+CD45RBhigh T cells from
AbwtAbEpIi- chimeras
and found that only the last subpopulation of CD4+ T cells
proliferates when cultured on plates coated with anti-TCR mAb (data
not shown). Furthermore, we analyzed activation markers expressed on
CD4+ T cells from newly established
AbEpIi-ß2m- mice. In contrast to
the original AbEpIi- mice, the
AbEpIi-ß2m- mice had majority of
CD4+ T cells CD44lowCD69low
CD45RBhigh (Fig. 3). We concluded that the repertoire of
CD4+ T cells in AbEpIi-
mice and AbwtAbEpIi-
chimeras includes a number of CD4+ T cells selected on
nonclassical MHC elements, and these T cells often express activation
markers as previously described (40). In contrast, most of
CD4+ T cells positively selected on the AbEp
complex are similar to naive CD4+ T cells in wild-type mice
and can normally respond to challenge with Ags (see below).
|
AbEpIi- chimeras
changed the usage of different TCR Vß segments in the
CD4+ T cell repertoire. For this purpose, the
CD4+ T cells selected in
AbwtAbEpIi- chimeras
were stained with a set of mAbs specific for various TCR Vß segments
and compared with the Vß repertoire of CD4+ T cells
selected in
AbEpIi-AbEpIi-
and AbwtAbwt chimeras. As shown in Fig. 4A, introduction of the
negative selection on self peptides in
AbwtAbEpIi- chimeras
reduced the numbers of Vß4-, Vß12-, and Vß14-bearing
CD4+ T cells and increased the frequency of Vß5 and
Vß9. Furthermore, we have screened the TCR Vß repertoires expressed
by populations of CD4+CD45RBhigh T cells
from the same set of the radiation chimeras (Fig. 4B). This
staining revealed that the repertoire of
CD4+CD45RBhigh cells from
AbwtAbEpIi- chimeras
had reduced numbers of Vß4 and Vß14 and increased numbers of Vß5,
Vß7, Vß8, and Vß9 CD4+ T cells.
|
CD4+ T cells selected by single class II MHC peptide and tolerant to wild-type peptides preserved a broad repertoire of Ag specificities
One of the important questions concerning the formation of the natural repertoire of T cells is whether the selecting peptide ligands leave an imprint on the selected repertoire of TCRs such that the encoded Ag specificities are largely similar to those of the peptides that induced their positive selection (38). Likewise, it has been postulated that a diversity of self peptides is critical for positive selection in the presence of negative selection on many self MHC/peptide complexes to select TCRs with various Ag specificities (29). Mice that express single AbEp complex and have been reconstituted with wild-type fetal liver cells provide a model to investigate the Ag specificities of TCRs positively selected by the defined MHC/peptide ligand and tolerant to a normal set of self-derived peptides. The CD4+ T cell repertoire in such chimeras also represents the potential contribution of positive selection on a single class II MHC/peptide complex in the normal repertoire of CD4+ T cells.
Subsequently, we performed proliferation assays to test whether a
repertoire of CD4+ T cells selected on a single
AbEp complex and tolerant to self peptides is skewed toward
preferential interaction with Ags that are related in primary amino
acid sequence to the selecting Ep (Fig. 5). For this purpose,
AbwtAbEpIi- chimeric
mice were separately primed with several antigenic peptides. The
CD4+ T cells were isolated from these immunized chimeras as
described and incubated with the same antigenic peptides presented by
the irradiated spleen APCs from wild-type mice. On days 3 and 5, IL-2
was added to further expand cells that specifically responded to the
added peptide, and the proliferative response was measured on day 7 as
previously described (41).
|
. Two of them are single amino acid
analogues of Ep52–68 and were chosen to represent
antigenic peptides that are similar in primary amino acid sequence to
the selecting Ep. These two Ep analogues were frequently recognized by
CD4+ T cells from
AbEpIi- mice, and changed amino
acids in position 58 or 63 have been mapped as TCR contact residues
(30). Two other peptides different from Ep were also included in this
test. One of these (EpKKK) was designed on the backbone of Ep with
three TCR-facing residues changed to lysine. The fourth antigenic
peptide tested, PCC43–58, was chosen from a set of
peptides known to bind to Ab molecules.
|
, the repertoire of CD4+ T cells selected
on AbEp complex and tolerant to wild-type Ab
complexes did not respond to the tested Ep analogues significantly
better than to the rest of the tested peptides. For example, one of the
Ep analogues with single substitution in TCR contact residues did not
trigger an enhanced CD4+ T cell response. Moreover, in
proliferation assays the PCC43–48-derived peptide was a
very good stimulator of CD4+ T cells selected in
AbwtAbEpIi- chimeras,
and this peptide differs significantly from Ep.
In conclusion, the finding that peptides very different from the
selecting Ep induced proliferation of CD4+ T cells selected
in Abwt AbEpIi-
chimeras showed that this repertoire, although quantitatively
restricted, preserves the capacity to respond to various Ep-related and
unrelated Ags. This result contradicts the presumption that the
original, quite diverse spectrum of CD4+ T cells found in
AbEpIi- mice will encode few Ep-like
Ag specificities after negative selection in vivo on wild-type
Ab/peptide complexes (29, 38).
Repertoires of TCRs specific for the same Ag and selected in
AbwtAbEpIi- chimeras or
wild-type mice can be similar, but the former repertoire is more
restricted
To further investigate the composition of TCRs selected by
AbEp complex and tolerant to self-derived peptides, we
again primed the
AbwtAbEpIi- chimeric
mice with six different soluble peptides in the presence of CFA (their
amino acid sequences are listed in Table I). The goal of these
experiments was to isolate a set of Ag-specific TCRs that developed in
these chimeras and to determine their primary sequences. At the same
time we primed wild-type mice with identical antigenic peptides so that
the wild-type repertoire of TCRs specific for the same antigenic
peptides was generated. CD4+ T cells from Ag-primed
chimeras and wild-type mice were converted into T cell hybridomas,
followed by screening for CD4/TCR expression and Ag specificity. As
expected, a large number of CD4+ T cell hybridomas specific
for each tested peptide was produced from wild-type mice. The number of
CD4+ T cells in
AbwtAbEpIi- chimeras
is only 3–5% of that seen in the wild-type mice, although in two
subsequent immunizations for each tested peptide, we generated a total
of 25 CD4+ T cell hybridomas specific for different
antigenic peptide. This result correlated with the proliferation assay
data and confirmed that the repertoire of CD4+ T cells
selected by single AbEp ligand and tolerant to wild-type
Ab/peptide complexes preserves TCRs with various Ag
specificities. Interestingly, peptide PCC43–58, which was
the most unrelated in primary sequence to Ep, was found to be
recognized by five different TCRs that originated from
AbwtAbEpIi- chimeras.
Moreover, one CD4+ T cell hybridoma specific for EpKKK and
one for IgGvH59–74 peptide were also isolated (see below).
To compare the complexity of TCR repertoires of CD4+ T cell
hybridomas specific for the same antigenic peptide from
AbwtAbEpIi- or
wild-type mice, we determined their Vß repertoires using flow
cytometry. As expected, the repertoire of TCRs specific for every
tested antigenic peptide was significantly more diverse in the
hybridomas derived from wild-type mice than in those from
AbwtAbEpIi- chimeras.
Despite differences in the numbers of CD4+ T cells, both
repertoires often shared the most prevalent Vß segments. For example,
either Vß6 or Vß14 subpopulations were commonly present among
CD4+ T cell hybridomas specific for Ep58K or Ep63K
analogues from both types of mice (Fig. 6). In addition, as was previously
reported for AbEpIi- and wild-type
mice, four of six hybridomas specific for PCC43–58 were
Vß8+ (30).
|
AbEpIi- chimeras
cross-react with different Ep analogues. The intent of this experiments
was to evaluate the hypothesis that Ab molecules
exclusively bound with Ep may positively select CD4+ T
cells with a permissive mode of Ag recognition (38). In general, we
found that isolated T cell hybridomas are peptide specific and that
some can discriminate a single amino acid substitution in antigenic
peptide. For example, three different TCRs selected in AbEp
exclusively responded to Ep58K, but not to Ep58R or any other tested
peptides (Fig. 7). Similarly, of five
TCRs specific for the Ep analogue with K and/or R in position 63
(Ep63K, Ep63R), four showed no response to other tested analogues of
Ep. Obviously, some of the tested T cell hybridomas shared a common
epitope, but no common propensity toward Ep-like Ags could be found. We
concluded that TCRs selected in
AbwtAbEpIi- chimeras
have no prevalent Ep reactivity and are Ag specific, as are normal
CD4+ T cells from wild-type mice.
|
AbEpIi- chimeras is
diverse, but some TCRs are similar
Since the repertoire of CD4+ T cells positively
selected on single AbEp complex and tolerant to Abwt was
found to encode multiple Ag specificities, we further investigated the
primary DNA sequences of these TCRs. The aim of this analysis was to
determine whether the TCRs expressed on CD4+ T cells in
AbwtAbEpIi- chimeras
that are capable of responding to a given peptide Ag are related.
The complementarity-determining region (CDR) sequences of 25
Ag-specific TCRs from
AbwtAbEpIi- chimeras
were determined (Fig. 8). Two-thirds of
these TCRs were specific for the tested Ep analogue. Six other TCRs
were specific for PCC43–58, one for
IgGvH59–74, and one for Ep(KKK).
|
-chain with the same amino
acid sequence coupled with three different TCR ß-chains. As shown in
Table I, the PCC43–58 peptide and Ep58K analogue are quite
different, albeit the p3p5 residues in both peptides may be acceptable
for the same TCR ß-chain if the proposed common orientation of TCR
interacting with class II MHC/peptide complex is correct (42).
Furthermore, one TCR (no. 64) specific for IgGvH59–74 and
one (no. 108) specific for Ep58K analogue express a similar combination
of Vß2Jß2.5(or 2.2)/V10. Likewise, two
PCC43–58-specific TCRs (no. 15 and 2.1) and one (no. 16)
specific for soluble Ep had a similar pattern of V4/Vß8.1Jß2.6.
The fact that these similar TCRs were found within two dozen TCRs
derived from separate priming experiments suggests that the repertoire
of CD4+ T cells in
AbwtAbEpIi- chimeras
is compressed.
The similarities described above between TCRs with different Ag
specificities were even more profound among TCRs specific for the same
antigenic peptide. For example, the majority (three of five) of all
PCC-specific TCRs had a Jß2.6 segment. This original observation made
for PCC-specific TCRs selected in
AbEpIi- mice also appears to apply
to TCRs selected in
AbwtAbEpIi- chimeras
(30). Moreover, all PCC-specific TCRs used V4 segments exclusively.
The family of V4 genes encodes 12 known variable segments, of which
3 (V4.1, V4.5, and V4.12) were used by PCC-specific TCRs
selected in
AbwtAbEpIi- chimeras.
The former two V4 segments have nearly identical CDR1 and CDR2
regions (Fig. 8). We also found that two PCC43–58-specific
hybridomas express identical DNA sequences V4.1J8/Vß8,1Jß2.6
TCRs, so it is most likely they both originated from the same
CD4+ T cell clone. Finally, the length of the CDR3 region
among all PCC-specific TCRs varied by only one amino acid in both TCR
- and ß-chains. In contrast, the length of CDR3 in TCRs specific
for the two tested Ep analogues was usually three amino acids in both
TCR chains as described below (see Fig. 8). This result indicates that
a single class II MHC/peptide complex may have more versatility for
positive selection of some TCRs.
The rest of the TCRs isolated from
AbwtAbEpIi- chimeras
were specific for different Ep analogues. These TCRs were resorted into
six groups based on their reactivities with different Ep analogues (see
Fig. 7).
The first group includes TCRs that specifically responded to Ep
analogues with K or R present on the C' terminus of the antigenic Ep
analogue. Four of five TCRs specific for these large, charged, side
chains in position p8 of Ep used either Vß6 or Vß14 segments. We
found no other similarities in V usage among these TCRs. The fifth
TCR (no. 48) in this category responded not only to Ep63K and Ep63R
analogues but also to EpKKK. This last TCR was originally generated by
priming AbwtAbEpIi-
chimeras with EpKKK, which probably explains its further relationship
to the rest of the TCRs in this group.
The second group included three TCRs exclusively specific for a single
large charged substitution in position p2 of Ep. Each of these TCRs
used different TCR - and ß-chains, and even the lengths of the
CDR3 regions between these TCRs are different. However, as mentioned
above, two of these TCRs are similar to TCRs specific for other tested
antigenic peptides (Fig. 8).
The third group included TCRs that responded to Ep analogues with an
unchanged middle sequence of the peptide, in particular the p5
position. Remarkably, two of these TCRs (no. 85 and 141.5) express
completely different TCR -chains coupled with almost identical TCR
ß-chains (one amino acid difference in CDR3 region). Since the p5
position is likely to contact CDR3 regions on both TCR chains, the
similarity between these two TCR ß-chains is being further
investigated.
The fourth group of TCRs included hybridomas that responded to all the tested Ep analogues that lacked large side chains in the N terminus of the Ep sequence. These TCRs did not respond to Ep analogues with alanine substituted by lysine in the p2p5 positions of Ep. Three TCRs in this group were very different, although one was noticed to be similar to one of the TCRs from group 6.
Two TCRs in group 5 responded to Ep analogues that lacked a substantial
side chain in position p2 and were not stimulated by Ep analogues
expressing K or R at this position. Perhaps the CDR1 and CDR2 regions
encoded by the V genes in these two TCRs could not accommodate such
substantial changes. No particular homology with other TCRs was
noticed.
Two TCRs that could not be placed in any of the five other groups were
put into group 6. One of these (no. 16) was exclusively specific for
soluble Ep (see Fig. 7). Currently, we are investigating why other Ep
analogues were not stimulatory for this TCR. The other TCR in this
group (no. 94) responded to every tested Ep analogue with a single
amino acid substitution.
In conclusion, Ag-specific TCRs isolated from
AbwtAbEpIi- chimeras
revealed that this repertoire remains relatively diverse but far less
heterogeneous than the wild-type repertoire of TCRs specific for the
same Ags, as immediately evident in a comparison of the TCR Vßs
repertoires (Fig. 8). Most likely, positive selection on the single
AbEp complex combined with extensive negative selection on
the normal spectrum of self peptides forces this TCR repertoire to use
a more limited spectrum of V, D, and J TCR segments. TCRs specific for
antigenic peptides unrelated to Ep seemed to be more restrained than
those specific for Ep analogues.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/Vß segments able to respond to Ags unrelated to the selecting
peptide. This finding also implies that poor shape complementarity
found in crystal structure of TCR engaged by self class I MHC/peptide
complexes is likely to be a common mode for in vivo interactions that
provoke thymic positive selection of majority of thymocytes (43).
However, it would be misleading to assume that one class II MHC/peptide
element can substitute for MHC molecules bound with many self-derived
peptides in selecting T cells in the thymus. There are visible
limitations, and probably unbalances, in the representation of various
Ag specificities in a repertoire selected by a single class II
MHC/peptide complex. Despite dramatic quantitative differences (only
3–5% of the number wild-type CD4+ T cells can be found in
AbwtAbEpIi- chimeras), these
CD4+ T cells responded to all tested antigenic peptides
with a far less diverse set of TCRs; this was easily detected by
staining Ag-specific TCR Vß segments (Fig. 7). Amino acid sequences
of Ag-specific TCRs revealed that this repertoire contains some similar
TCRs able to mount specific responses to Ep-related and unrelated Ags.
For six tested antigenic peptides, two-thirds of the isolated TCRs were
specific for peptides related to Ep. This may imply that the repertoire
may be enriched in some Ag specificities. Simultaneously, however, we
did not record a significant bias in proliferative responses to Ep-like
Ags, which suggests that potential skewing in this repertoire is not as
significant as previously suggested (40). Likewise, an unbalanced
distribution of Ag specificities of CD4+ T cells in
AbEpIi- or
DM- mice was not recorded in previous in vivo
experiments (28, 30, 39).
The majority of CD4+ T cells positively selected in vivo on
single AbEp complexes are negatively selected on
Ab molecules occupied with endogenous peptides. The extent
of the observed deletion, approximately two-thirds, is the same for
repertoires positively selected on two unrelated peptides bound to
Ab molecules (Ep or CLIP) expressed at either low or normal
levels (Ref. 28 and this report). Here we have shown that this profound
deletion reflects that lack of tolerance to self peptides presented in
the context of the selecting II MHC. It is unlikely that positive
selection by a single MHC/peptide complex narrows such a repertoire
toward one or a few Ags. For instance, some of the AbEp
selected TCRs that are subject to negative selection on wild-type
Ab/peptide complexes, as tested by in vitro response to
Abwt APCs alone, also responded to different antigenic
peptides (30). In addition, in
AbwtAbEpIi- chimeras,
CD4+ T cells remaining after negative selection are able to
mount specific responses against several very different antigenic
peptides. Therefore, the TCR repertoire generated by positive selection
on the AbEp complex not only exceeds negative selection on
the same AbEp, but also overcomes negative selection on
many other self MHC/peptide complexes and contributes a noticeable
number of different CD4+ T cells to the periphery. Since
positive selection on AbEp complex promotes 3–5% of the
normal number of CD4+ T cells to the periphery, potentially
around 100 different self class II MHC/peptide complexes may positively
select almost a normal number of CD4+ T cells, a published
hypothesis that has not been experimentally tested (44).
Around 50 and 80% of peripheral CD4+ T cells in
AbEpIi- and
AbwtAbEpIi- mice are
CD45RBlowCD44high, respectively. In contrast,
<20% of these CD4+ T cells are found in wild-type or
AbEpIi-ß2m- mice. Moreover, none
of the mice tested had many CD45RBlowCD44high
SP CD4+ thymocytes, as previously reported for
DM- chimeric mice. These findings suggest that
many of these CD4+ T cells may be selected on nonclassical
MHC molecules, and they can be abundant in the periphery of mice with
restricted positive selection on class II/peptide complexes. Therefore,
quantitative and qualitative analysis of the repertoire of TCRs
selected by the single or dominant class II MHC/peptide complex will be
adequate after this type of CD4+ T cells is excluded.
Meanwhile, in
AbEpIi-ß2m- mice,
the absolute number of peripheral CD4+ T cells with a naive
phenotype almost triples, indicating that the peripheral pool of
AbEp selected CD4+ T cells is also controlled
by unknown homeostatic mechanisms (see Fig. 3).
Introduction of the negative selection on wild-type Ab
peptide complexes in our chimeric
AbwtAbEpIi- mice also
removed an abundance of alloreactive CD4+ T cells present
in AbEpIi- mice. We have tested 50
CD4+TCR+ T cell hybridomas from
AbwtAbEpIi- chimeras
and found them unresponsive to three different class II MHC molecules
(data not shown). This result agrees with our former postulate that
negative selection on many self-derived peptides calibrates the MHC
restriction such that immature T cells excessively reactive to MHC per
se are largely eliminated in the thymus (22).
Analysis of TCRs selected on AbEp and tolerant to wild-type self peptides indicates that this repertoire is relatively heterogeneous, with no dominant pattern of Ag reactivity. The diversity of CDR regions of TCRs specific for Ep-like Ags is higher then the diversity of CDRs of TCRs specific for Ep-unrelated peptides such as PCC. Since covalently bound Ep probably projects one substantial side chain, an isoleucine at position p8 toward TCR, one may argue that peptides with more "bulky" TCR contact residues will positively select a narrower, but more specific, set of TCRs (45). Future experiments involving analysis of Ag specificities and CDR sequences of TCRs selected on MHC coupled with flat or "hairy" peptides should reliably test this speculation.
What, then, are the general rules that govern the positive and negative selection of T cells in the thymus? It seems that positive selection is "promiscuous with borders" such as TCRs with the same Ag specificities are selected by various MHC/peptide ligands. This property of thymic selection may ensure that the specific Ag response does not entirely depend on positive selection by one self MHC/peptide complex. Likewise, for many TCRs there is more than one self MHC/peptide combination able to promote their positive selection. Finally, since the same self peptides are likely to be involved in positive and negative selections, the repertoire that leaves the thymus has scattered, rather then narrow, specificities. Such a general mode of thymic selection, where positive selection is less peptide dependent than negative, enhances its effectiveness and increases the diversity of T cells specific for a particular Ag.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Leszek Ignatowicz, Institute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, GA 30912-2600. E-mail address:
3 Abbreviations used in this paper: FTOC, fetal thymus organ culture; CLIP, class II-associated invariant chain peptide; Ii, invariant chain; PCC, pigeon cytochrome c; BSS, balanced salt solution; MTT, 3-(4,5-dimethythiazol-2-yl)-2,5-diphenyltetrazolium bromide; CDR, complementarity-determining region.
Received for publication May 27, 1998. Accepted for publication September 3, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. L. Stephens, J. D. Ashwell, and L. Ignatowicz Mutually antagonistic signals regulate selection of the T cell repertoire Int. Immunol., May 1, 2003; 15(5): 623 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Pacholczyk, P. Kraj, and L. Ignatowicz Peptide Specificity of Thymic Selection of CD4+CD25+ T Cells J. Immunol., January 15, 2002; 168(2): 613 - 620. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Peggs, S. Verfuerth, A. Pizzey, J. Ainsworth, P. Moss, and S. Mackinnon Characterization of human cytomegalovirus peptide-specific CD8+ T-cell repertoire diversity following in vitro restimulation by antigen-pulsed dendritic cells Blood, January 1, 2002; 99(1): 213 - 223. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Kraj, R. Pacholczyk, and L. Ignatowicz {{alpha}}{{beta}}TCRs Differ in the Degree of Their Specificity for the Positively Selecting MHC/Peptide Ligand J. Immunol., February 15, 2001; 166(4): 2251 - 2259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Pacholczyk, P. Kraj, and L. Ignatowicz An Incremental Increase in the Complexity of Peptides Bound to Class II MHC Changes the Diversity of Positively Selected {{alpha}}{{beta}} TCRs J. Immunol., February 15, 2001; 166(4): 2357 - 2363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gaszewska-Mastalarz, P. Muranski, B. Chmielowski, P. Kraj, and L. Ignatowicz Altered Selection of CD4+ T Cells by Class II MHC Bound with Dominant and Low Abundance Self-Peptides J. Immunol., December 1, 2000; 165(11): 6099 - 6106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Muranski, B. Chmielowski, and L. Ignatowicz Mature CD4+ T Cells Perceive a Positively Selecting Class II MHC/Peptide Complex in the Periphery J. Immunol., March 15, 2000; 164(6): 3087 - 3094. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Chmielowski, P. Muranski, P. Kisielow, and L. Ignatowicz On the role of high- and low-abundance class II MHC-peptide complexes in the thymic positive selection of CD4+ T cells Int. Immunol., January 1, 2000; 12(1): 67 - 72. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
