| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Lymphocyte Biology Section, Division of Rheumatology, Immunology, and Allergy, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115;
Department of Molecular Biology and Skaggs Institute of Chemical Biology, The Scripps Research Institute, La Jolla, CA 92037; and
Department of Microbiology and Immunology, School of Microbiological, Immunological, and Virological Sciences, University of Newcastle upon Tyne, The Medical School, Newcastle upon Tyne, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TCR can recognize peptides presented by MHC molecules or
lipids and glycolipids presented by CD1 proteins. Whereas the
structural basis for peptide/MHC recognition is now clearly understood,
it is not known how the TCR can interact with such disparate molecules
as lipids. Recently, we demonstrated that the TCR confers
specificity for both the lipid Ag and CD1 isoform restriction,
indicating that the TCR is likely to recognize a lipid/CD1 complex. We
hypothesized that lipids may bind to CD1 via their hydrophobic alkyl
and acyl chains, exposing the hydrophilic sugar, phosphate, and other
polar functions for interaction with the TCR
complementarity-determining regions (CDRs). To test this model, we
mutated the residues in the CDR3 region of the DN1 TCR -chain that
were predicted to project between the CD1b helixes in a model of
the TCR/CD1 complex. In addition, we tested the requirement for the
negatively charged and polar functions of mycolic acid for Ag
recognition. Our findings indicate that the CDR loops of the TCR form
the Ag recognition domain of CD1-restricted TCRs and suggest that the
hydrophilic domains of a lipid Ag can form a combinatorial epitope
recognized by the TCR. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and -chains of the TCR
heterodimer consist of Ig domains, with the highly variable
complementarity-determining region
(CDR)8 loops forming a
surface that interacts with the MHC/peptide complex
(1, 2, 3, 4, 5). MHC molecules contain a peptide-binding groove
comprised of multiple pockets with binding specificity for backbone and
particular amino acid side chains, enabling peptides with appropriate
residues at critical anchor positions to bind with high affinity
(6). This mechanism of binding allows other amino acid
residues in the MHC-bound peptide to orient toward the TCR interface
where direct interactions with the TCR CDR loops occur, thereby
contributing to the affinity and specificity of the interaction. In the
TCR/peptide/MHC structures determined to date, the TCR adopts a range
of diagonal toward orthogonal orientations relative to the MHC
helixes, positioning the TCR CDR3 loops centrally over the peptide and
the CDR1 and -2 loops peripherally, where interactions with the MHC
helixes and the ends of the bound peptides can occur (1, 2, 5). This high precision interaction drives T cell selection in
the thymus and is critical for T cell responses in the periphery
against microbial infections, including inflammatory and cytolytic
responses, and for induction of Ig production by B cells.
Recently, we and others demonstrated that TCRs can recognize
foreign lipid Ags in the context of CD1a, -b, -c, or -d presenting
molecules, supporting a role for Ag presentation by CD1 in microbial
immunity (7, 8, 9, 10, 11, 12). Yet, the molecular basis by which the
TCR might recognize lipid Ags is only beginning to be
understood. Given that the lipid Ags identified to date contain a
hydrophilic head and a hydrophobic tail, we proposed that CD1 molecules
would bind the lipid tails and that the hydrophilic head would be
recognized by the TCR (13). This hypothesis is now
supported by the finding that the CD1d atomic structure reveals an
Ag-binding superdomain that contains two electrostatically neutral
pockets (A' and F') lined predominantly by nonpolar and hydrophobic
residues, consistent with the ability to bind lipid tails
(14). Although no structure is yet available for a CD1/Ag
or a CD1/TCR complex, we recently proposed a molecular model based on
the overall topological similarity between CD1 and MHC proteins
(15). This model predicted that the TCR CDR3 loops would
be positioned over the opening to the hydrophobic cavity located
between the helixes at the top of the CD1b extracellular domain.
Importantly, the TCR -chain CDR3 loop of the DN1 TCR contains a pair
of basic arginine residues positioned directly over the entrance
to this hydrophobic pocket. Since no neighboring residues of the CD1b
protein are similarly hydrophilic, we suggested that these TCR CDR3
loop basic residues might be involved in direct interactions with the
negatively charged carboxylic acid portion of mycolic acid, the
CD1b-restricted Ag recognized by the DN1 TCR (15).
Mycolic acids are a prototypical foreign microbial lipid Ag, since they
are the predominant lipid component of the mycobacterial cell wall but
have no counterpart in mammals. They are found as esters of
arabinogalactan and trehalose, where they are covalently linked to
carbohydrates as in glucose monomycolate (GMM) and as free fatty acids
that contain a long (approximately C60) main
alkyl chain with a second alkyl chain (C20 to
C25) attached to the carbon and a hydrophilic
hydroxy function at the carbon, and are thus referred to as
-branched, -hydroxy fatty acids (16, 17). In
Mycobacterium tuberculosis, three types of mycolic acids are
present that are divided into subclasses, referred to as , methoxy,
and keto based on the presence of either a cyclopropanoid, methoxy, or
keto function, respectively, on the main (longer) alkyl chain
(16, 17). Recent studies have identified residues in the
CD1b Ag binding groove, across both the A' and F' pockets, that are
critical for presentation of GMM, implying a role in binding the alkyl
chains of mycolates (18). Other studies point to surface
helical residues of CD1b that might be predicted to interact
directly with the TCR, consistent with a diagonal footprint of the TCR
across the long axes of the CD1 helixes (19). Yet no
studies have critically examined the structural basis for recognition
by the TCR.
To determine the role of TCR CDR3 residues in the recognition of lipid
Ags by an TCR, we performed site-directed mutagenesis of the
arginine residues located at the tip of the TCR CDR3 loop and
tested the ability of TCR-transfectant cell lines to be activated by
mycolic acid presented by CD1b-bearing APCs. Using this approach we
mapped the TCR CDR3 residues necessary for recognition of the fatty
acid Ag. Correspondingly, we determined that the polar substitutions (R
groups) attached to the mycolic acid main alkyl chain and a free
carboxylic acid group were necessary for recognition of this lipid Ag
by the DN1 TCR.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The J.RT3-T3.5 cell line (American Type Culture Collection,
Manassas, VA), a derivative of Jurkat that has a defect in the
expression of endogenous TCR -chains (20), was
maintained in RPMI 1640 supplemented with 10% FCS, HEPES,
L-glutamine, 2-ME, and penicillin/streptomycin (complete
medium). The IL-2-dependent cell line HT-2 (American Type Culture
Collection) was maintained in complete medium further supplemented with
2 nM rhIL-2, essential and nonessential amino acids, and sodium
pyruvate. The following mAbs were used for analysis of cell surface TCR
and CD3 levels: SPVT3b (anti-CD3
, IgG2a) (21) and
F4#120K.1 (anti-DN1 TCR clonotype, IgG1). The mAb F4#120K.1 was
generated from a mouse immunized with DN1 T cells and isolated on the
basis of its inhibition of DN1 T cell proliferation in culture with
Ag-loaded APCs as previously described (22). The mAb
appears to recognize a combinatorial epitope formed by the DN1 TCR -
and -chains based on flow cytometric analyses of PBL and J.RT3-T3.5
cells transfected with one or both chains of the DN1 TCR (data not
shown).
Mutagenesis
The wild-type DN1 TCR (TCRAV8S2AJ57) and TCR
(TCRBV5S1J2S7) cDNAs were previously cloned and ligated into the pREP7
and pREP9 expression vectors (Invitrogen, San Diego, CA), respectively
(15). For mutagenesis, the DN1 TCR cDNA was subcloned
into the pALTERMAX plasmid (Stratagene, La Jolla, CA), and mutagenesis
was conducted according to the manufacturer’s instructions. The
following mutagenic primers were used: R97A,
5'-gccagcagcttggtcgcgcgctacgagcagtacttc-3'; R98A,
5'-tactgctcgtaggccctgaccaagct-3'; R97K,
5'-ctcgtagcgcttgaccaagct-3'; R98K,
5'-gtactgctcgtatttcctgaccaagct-3'; R30A,
5'-gggttggtaccaggatacactcgcatgcccagagatagg-3'.Mutant plasmids
were subcloned into the pREP9 plasmid (Invitrogen) for transfection.
TCR transfection
J.RT3-T3.5 cells were resuspended at 4 x
107 cells/ml in complete medium, and 0.3 ml of
the cell suspension was transferred into each electroporation cuvette
(Bio-Rad, Hercules, CA). Twenty micrograms of pREP7-DN1 TCR and 20
µg wild-type or mutant pREP9-DN1 TCR plasmids were mixed with
cells. After a 10-min incubation at room temperature the cells were
electroporated at 250 V and 960 µF in a GenePulser electroporator
(Bio-Rad). The cells were incubated for 10 min at room temperature,
resuspended in complete medium, and cultured at 37°C. After 48
h, the cells were transferred to complete medium containing G418 (1
mg/ml; Life Technologies, Grand Island, NY) and hygromycin B (0.5
mg/ml; Calbiochem, La Jolla, CA). Transfectants were maintained in
selection medium for at least 2 wk before analysis by flow cytometry
and use in Ag presentation assays.
Cytofluorographic analysis
TCR transfectants were analyzed for cell surface expression of
TCR/CD3 complexes as follows. Cells were incubated with mAbs SPVT3b
(anti-CD3), F4#120K.1 (anti-DN1 TCR clonotype), or
isotype-matched control Ig (20 µg/ml in PBS/2% FCS) for 45 min on
ice. Cells were washed and incubated with FITC-labeled goat
anti-mouse IgG/M F(ab')2 for 45 min on ice.
Cells were washed and analyzed on a FACSort flow cytometer (BD
Biosciences, Franklin Lakes, NJ).
Mycolic acid preparations
Mycolic acids were isolated from M. tuberculosis strain H37Ra as previously described (15). To form methyl esters, mycolic acids (10 mg) were dried in a 15-ml glass conical tube under N2 gas. One milliliter of 0.8% NaOH, 3.39% tetrabutylammonium hydrogen sulfate in dH2O was added to the tube, and the mycolic acids were suspended by sonication in a water bath sonicator for 2–3 min. One milliliter of dichloromethane and 25 µl iodomethane were added to the tube and incubated for 30 min at room temperature with intermittent vortexing. The tube was centrifuged at 1500 rpm for 5 min, and the upper phase was removed and discarded. One milliliter of 1 N HCl was added to the lower phase, vortexed, and centrifuged. The upper phase was removed and discarded. One milliliter of dH2O was added to the lower phase, vortexed, and centrifuged. The upper phase was discarded, and the lower phase was evaporated under N2 gas. The mycolic acid methyl esters were resuspended in chloroform and stored at -20°C. To assess the formation of methyl esters, approximately 100 µg were spotted on a 0.2-µm silica TLC plate and developed five times in petroleum ether-ethyl acetate (94/6). The TLC plates were sprayed with 3% (w/v) cupric acetate, 8% phosphoric acid and baked at 150°C for 15 min. TLC analyses of the methyl esters used in this report revealed three major spots with the characteristic pattern for mycolic acid methyl esters from M. tuberculosis (data not shown).
Ag presentation assay
Ag presentation assays were performed as previously described (15). Briefly, TCR/J.RT3 transfectants (105/well) were incubated in 96-well flat-bottom microtiter plates with monocyte-derived dendritic cells (GM-CSF/IL-4 treated peripheral blood monocytes) at 5 x 104/well together with the indicated concentrations of mycolic acids or mycolic acid methyl esters. PMA was added at a final concentration of 10 ng/ml in all wells. After 24 h incubation at 37°C, 25-µl aliquots of the supernatants were transferred to 96-well plates containing 75 µl fresh complete medium. HT-2 cells were added (5000/well) and incubated for 24 h at 37°C. [3H]Thymidine (1 µCi/well, 6.7 Ci/mmol; New England Nuclear, Boston, MA) was added during the final 5 h of incubation. Plates were harvested on a Tomtec 96-well plate harvester (Wallac, Gaithersburg, MD), and [3H]thymidine incorporation was determined using a Betaplate liquid scintillation counter (Wallac).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TCR chains from the mycolic
acid-specific, CD1b-restricted T cell clone DN1.F9 (15).
Transfection of these cDNAs into TCR-deficient J.RT3-T3.5 cells yielded
a transfectant line designated DN1/J.RT3 that produces IL-2 in response
to CD1b-expressing cells pulsed with mycolic acid, but not other lipid
Ags. This response was CD1b dependent, since the anti-CD1b mAb
BDC1b3.1 substantially blocked this response, while mAbs specific for
other CD1 isoforms, CD1a and CD1c, had no effect. Thus, the expression
of an TCR is sufficient to confer specificity for both a lipid
Ag and CD1 isoform restriction.
Computer modeling suggested that the DN1 TCR and CD1b could interact in
a complex similar to that between the TCR and MHC class I. This
DN1/CD1b model accommodates a diagonal orientation of the TCR V domains
across the helixes of CD1b, much like the orientation of MHC class
I-restricted TCRs to MHC-presenting molecules in the known crystal
structures (1, 2, 3, 4). In this orientation the CDR3 loops of
the DN1 TCR would be positioned over the opening to the hydrophobic
cavity between the CD1b helixes where the alkyl and acyl chains of
lipid Ags are predicted to bind. We have suggested that the hydrophobic
tails of lipid Ags may mediate binding in the CD1 groove, leaving the
hydrophilic moieties of the lipid Ags (i.e., carboxylate, phosphate,
hydroxyl, or glycan) exposed for direct interaction with the TCR. In
our model of the DN1/CD1b complex, the side chains of R97 and R98 in
the CDR3 loop extend prominently between the CD1 1 and 2
helixes toward the putative lipid Ag-binding site (Fig. 1
A, V-encoded dark blue
loop 3; Fig. 1B, note line connecting side chains R97 and
R98 to the binding groove between the CD1 helixes). We previously
suggested that these TCR CDR3 basic residues may form important
contacts with the exposed hydrophilic and negatively charged moieties
of mycolic acids. In turn, the CDR1 and CDR2 loops of the DN1 TCR are
predicted to make contacts with the CD1b helixes themselves through
interactions that may act to stabilize and orient the TCR diagonally
over the CD1b/lipid Ag complex (Fig. 1B, note lines
connecting TCR V CDR2 residue R50 to CD1 residue E156 and TCR V
CDR1 residue R30 to CD1 residue E80).
|
The computer model for the DN1 TCR/CD1b complex indicated that a
favorable orientation for the DN1 TCR -chain CDR3 loop sequence,
LVRRYEQY, would be positioned centrally over the groove between the CD1
helixes. The side chains of two arginine residues, R97 and R98,
located in the center of this loop were predicted to extend prominently
into the opening of this putative lipid Ag binding site (Fig. 1
). Thus,
to determine whether the CDR loops of the TCR form a lipid Ag/CD1b
combining site analogous to peptide/MHC-specific TCRs, we initially
selected these two residues, R97 and R98, for mutagenesis to alanine
residues (R97A and R98A, respectively; Fig. 2). TCR transfectant J.RT3-T3.5 cells
were produced expressing wild-type DN1 TCR, R97A TCR, or R98A
TCR together with the wild-type DN1 TCR. Flow cytometric analyses
of untransfected, TCR-deficient J.RT3-T3.5 cells revealed very weak
staining with the anti-CD3 mAb SPVT3b (15–16% positive) and with
mAb F4#120K.1 against the DN1 TCR clonotype (1–2% positive; Fig. 3). The weak staining of J.RT3-T3.5 cells
by SPVT3b presumably reflects low level expression of endogenous Jurkat
TCRs. After transfection of expression vectors containing the
wild-type DN1 TCR - and -chains, wild-type TCR DN1/J.RT3 cells
expressed substantial levels of TCR at the cell surface (71–76% cells
positive; mean fluorescence intensity, 127–199) with
anti-CD3 mAb SPVT3b (Fig. 3, A and B) and
clearly stained positive, although at a modest level, with the
anti-clonotypic F4#120K.1 mAb (50–54% positive; mean fluorescence
intensity, 38–50; Fig. 3, A and B). R97A and
R98A mutant DN1/J.RT3 transfectants expressed similar levels of TCR and
the DN1 clonotype, in general within 10% of the levels and percentage
of cells positive as the wild-type transfectants analyzed in the same
experiment (Fig. 3).
|
|
transfectants (Fig. 4A). The failure of these
mutants to respond was not a consequence of secondary mutations in the
TCR chains, as confirmed by sequencing analysis (data not shown).
Further, the wild-type and mutant TCR transfectants expressed similar
levels of TCR as judged by flow cytometric analysis with anti-CD3
mAb SPVT3b and with the anti-clonotypic mAb F4#120K.1 (Fig. 3A). Therefore, the mutant TCRs were not deficient in
recognition of mycolic acid due to a reduction in the level of their
expression. Importantly, the mutant TCRs were capable of generating
signals to the transfectant cell lines, since stimulation of both
wild-type and mutant transfectants by the superantigen staphylococcal
enterotoxin C3 (SEC3) resulted in the ample production of IL-2 with
nearly superimposable dose-response curves for wild-type and mutant TCR
transfectants (Fig. 4B). Thus, these CDR3 mutants had a
profound defect in recognition of the CD1b-restricted lipid Ag, without
a discernable effect on global folding of the TCR or on the intrinsic
ability of the TCR to mediate signal transduction by CDR3
region-independent activation.
|
). TCR transfectant cell lines produced with these
mutations expressed CD3 and the DN1 clonotype at the cell surface at
comparable levels to the wild-type DN1/J.RT3 cell line (Fig. 3B). These mutants were tested for recognition of mycolic
acid and the superantigen SEC3 as described above. Remarkably, even
these conservative mutations completely abolished the response to
mycolic acid while having no effect on the response to SEC3 (Fig. 4, C and D). Therefore, it appears that the arginine
residues in the CDR3 loop of the DN1 TCR are crucial to the ability of
the TCR to recognize mycolic acid in the context of CD1b. Thus, the
conversion of an arginine residue to lysine, corresponding to the
removal of a single amino group, is a sufficient structural difference
to abolish the ability of the DN1 TCR to productively interact with the
mycolic acid/CD1b complex. Interestingly, substitution of a lysine with
an arginine in the peptide in the interface of the 2C
TCR-H-2Kb complex resulted in a
105-fold increase in biological potency, also
indicating that very conservative structural alterations can have a
large effect biologically (23). CDR1 loop mutagenesis
In contrast to the CDR3 loop arginines, which, based on the model,
we predicted might interact with Ag displayed at the opening of the Ag
binding groove between the CD1b 1 and 2 helixes, the model
suggested that residue R30 in the DN1 TCR CDR1 loop sequence,
ISGHRS, might make hydrogen bonds or a salt bridge to residue E80 on
the 1 helix of CD1b, thereby contributing to the binding energy of
the TCR/CD1b interface. Therefore, we produced a mutant TCR -chain
with alanine substituted for arginine at position 30 (R30A),
transfected it into J.RT3-T3.5 cells together with the wild-type DN1
TCR -chain, and analyzed the effect of the mutation on Ag
recognition. These R30A/J.RT3 transfectants were severely hindered in
their ability to produce IL-2 in response to mycolic acids, with a
response 20- to 40-fold lower than that of wild-type DN1/J.RT3 cells
(Fig. 5A). For comparison, the
R97A and R98A mutants showed no response to mycolic acid in the same
experiment. None of the mutants was substantially affected in
responsiveness to superantigen-mediated signaling, indicating the
expression of comparable levels of functional TCRs by all transfectants
(Fig. 5B).
|
Mycolic acids are long-chain (C60–C90), -branched, -hydroxy
fatty acids (16, 17). Unlike mycolic acids in other
species, those present in mycobacteria characteristically have
subclasses of mycolates with an R group, typically containing an oxygen
function on the longer (mero) chain at a position distal to the
carboxylate. This results in the mycolates of M.
tuberculosis being subdivided into those termed mycolates
(containing a cyclopropanoid unsaturation, but no R group oxygen
function), keto mycolates (containing an R group composed of a keto
moiety), or methoxy mycolates (containing a methoxy function; Fig. 6A). The mycolic acid
preparations used in the analyses described above were isolated by
saponification of the mycobacterial cell wall to release the mycolates
esterified to the covalent backbone of the cell wall as well as those
mycolic acids that are only hydrophobically associated with the cell
wall, i.e., trehalose mono- and dimycolate. Such a preparation
constitutes a total mycolate preparation from M.
tuberculosis, consisting of a mixture of these three classes of
mycolic acids. To test whether mycolic acid subtypes are differentially
recognized by the DN1 TCR, we separated the three classes by
preparative TLC based on their differential mobility in a petroleum
ether/ethyl acetate solvent system on silica plates. Purified ,
methoxy, and keto mycolates were then scraped from the plate, extracted
from the silica, and tested separately for the ability to stimulate
untransfected J.RT3-T3.5 cells or wild-type DN1/J.RT3 transfectant
cells. When dose-response analyses were performed, methoxy and keto
mycolates stimulated DN1/J.RT3 cells with similar dose-response curves,
whereas mycolates stimulated only a minimal response and were at
least 100-fold less potent than the other mycolate classes (Fig. 7B). These responses were
TCR-mediated, since the untransfected J.RT3-T3.5 cell line did not
respond to any of the mycolates tested (Fig. 7A). This
finding indicates that an oxygen function (keto or methoxy) on the
mycolic acid main chain (see Fig. 6) is necessary for recognition of
the CD1b-mycolic acid complex by the wild-type DN1 TCR. None of the
mutant TCRs tested responded to any of the mycolic acid subclasses,
indicating that the mutations abrogated recognition of both methoxy and
keto mycolic acids (data not shown).
|
|
, methoxy, and
keto mycolic acid methyl esters, in which a methyl group is covalently
linked via an ester linkage to the carboxylic acid function (see Fig. 6). These methyl esters were tested for antigenicity with wild-type
DN1/J.RT3 TCR transfectants. Whereas free mycolic acids were recognized
strongly, the mycolic acid methyl esters stimulated only background
levels of IL-2 production by DN1/J.RT3 cells (Fig. 7C).
Therefore, methylation of the carboxylate, which destroys its negative
charge by converting it to a methyl ester, abolishes the recognition of
mycolic acid by the DN1 TCR. Thus, the carboxylate is also necessary
for recognition by the DN1 TCR. A similar structural modification,
converting the free carboxylic acid to an ester linkage, occurs in
nature when glycosylated forms of mycolic acids, such as glucose
monomycolate or the glycosidically linked trehalose 6,6'-dimycolate
(cord factor), are synthesized by mycobacteria. Notably, these
compounds also are not recognized by DN1 T cells, in contrast to free
mycolic acid, which is recognized (data not shown). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cells to recognize either short peptides
or lipid and glycolipid Ags challenges us to explain in molecular terms
how these chemically distinct types of molecules can both be monitored
by TCRs encoded by the V and V genes. This enigma was partially
resolved with the appreciation that CD1 Ag-presenting molecules have an
overall topological similarity to MHC proteins, but with a cavity
consisting of two electrostatically neutral, hydrophobic pockets
located between the helixes that probably represent a binding site
for the alkyl and acyl chains of lipid and glycolipid Ags. Computer
modeling of the DN1 TCR as a complex with human CD1b predicted that the
and TCR CDR3 residues are positioned to interact with the
exposed portions of a bound Ag, such as the charged or polar regions
less likely to favorably interact with the hydrophobic CD1 groove
(15).
To test this model we explored the importance of the basic residues in
the DN1 TCR CDR3 loop and two of the three hydrophilic moieties of
the lipid Ag mycolic acid. Strikingly, we observed that alteration of
the central polar TCR CDR3 side chains had dramatic effects on
TCR/CD1/lipid recognition. In fact, even very conservative alterations
to the TCR CDR3, such as the replacement of arginine with lysine,
completely abrogated the ability of the TCR to respond to the mycolic
acid Ag. Similarly, we demonstrated that both the fatty acid
hydrophilic function on its long chain and the free carboxylic acid are
essential for the DN1 TCR to efficiently recognize this Ag. These
findings highlight the importance of both the TCR CDR3 sequence and the
precise molecular structure of the lipid Ag in the process of lipid Ag
presentation and recognition. Our results are consistent with the
proposed model for this complex and strengthen the hypothesis that the
TCR interacts with lipids and CD1-presenting elements via the
surface formed by the CDR1, -2, and -3 loops in an orientation
remarkably like that for interaction with MHC/peptide complexes. While
these data are consistent with our model of TCR/CD1b interaction, it
remains possible that the orientation of the TCR to the CD1 helical
domains could be different in terms of the angle and tilt of the TCR
relative to the putative lipid Ag-presenting domain. However, our
results clearly indicate that the surface formed by the TCR CDR loops
is, in fact, the Ag-combining site for a lipid-specific TCR and,
therefore, must interact with the CD1/lipid complex. A definitive
orientation is impossible to predict from the available data and must
await crystallographic determination of the structure of the
complex.
Mutation of residue R30 in the DN1 TCR -chain CDR1 loop reduced the
recognition of CD1b-presented mycolic acid. Based on our model, this
residue is predicted to hydrogen bond or form a salt bridge with
residue E80 of the CD1b 1 helix. Residue E80 lies at the entrance to
the Ag-binding pocket and corresponds to residue D80 in the mouse CD1d1
molecule. In the crystal structure of CD1d1, D80 is one of the rare
hydrophilic residues that comprise part of the Ag-binding groove and is
situated at the entrance to the groove. By extension, the CD1b residue
E80 may contribute to Ag binding in addition to its contribution to the
interface with the TCR.
Although the mutational analysis revealed striking effects on the
response to mycolic acid, none of the CDR1 or CDR3 mutants was
substantially impaired in response to the superantigen SEC3, which
probably binds to V5.1 (the V region expressed in the DN1 TCR)
through HV4, CDR2, and CDR1 loop interactions (24, 25, 26).
Further, flow cytometric analyses of the TCR transfectants using an
anti-clonotypic mAb indicated that similar levels of cell surface
expression of the DN1 TCR epitope were achieved for wild-type and
mutant TCRs. Taken together, these data indicate that the mutations of
the CDR3 loop specifically inhibited mycolic acid/CD1b recognition
without significantly altering the folding, expression, or signaling
potential of the mutated TCRs.
GMM, a glycosylated variant of mycolic acid is recognized by the
CD1b-restricted T cell line LDN5. Previous studies with this T cell
line demonstrated that the LDN5 TCR was highly sensitive to alterations
to the glucose component of GMM (27, 28). In contrast,
alterations to the alkyl chains of the Ag permitted recognition of the
Ag, although at reduced levels. These studies supported the idea that
the TCR interacts with the polar glucose moiety of that glycolipid Ag,
whereas the precise length of the lipid chains was not crucial, since
they probably mediate binding of GMM to CD1b rather than forming part
of the TCR epitope. The current studies examine recognition of the free
mycolic acids lacking an esterified hexose ring. The distinct
TCR and CD1-lipid complex examined here suggests that the DN1 TCR
interacts with two hydrophilic components of free mycolic acid.
Importantly, these components, the oxygen function present on the long
main (mero) chain and the carboxylic acid group at the opposite end of
the chain, are not adjacent. We suggest that these two hydrophilic
functions may form a combinatorial epitope when the alkyl chain folds
upon itself during binding to CD1b (Fig. 6B). This
combinatorial epitope may be positioned at the opening to the putative
CD1b Ag binding site, available for direct interactions with the
arginine residues in the TCR -chain CDR3 loop. We have not compared
the relative binding efficiencies of the various forms of mycolic acids
to CD1b. In addition, it is possible that the keto and methoxy
mycolates are internalized by APCs with greater efficiency than
mycolates or mycolic acid methyl esters. However, these alternate
explanations seem unlikely given that dramatic alterations in the
length of the mycolate chains present in the Ag GMM are permissive for
Ag presentation to LDN5 T cells (27). Further, GMM
molecules that contain or lack oxygen functions on the alkyl chains are
presented with similar efficiencies to the CD1b-restricted T cell line
LDN5, indicating that the absence of an oxygen function in
mycolates is not likely to prevent its internalization and presentation
by CD1b.
Although the extent to which lipid Ags are processed for presentation
by CD1 molecules is not known, recent data suggest that processing of
carbohydrate residues of lipid Ags can occur (29). It is
possible that the difference in recognition of keto and methoxy
mycolates compared with mycolates reflects differential processing,
in that mycolates contain an additional cyclopropyl group on the
mero chain, and this may influence processing of the lipid tail before
binding to CD1b. There are no known enzymes that might mediate this
processing function, but the possibility remains that the lipid tails
are covalently modified within an APC.
The ability of the TCR to perceive a specific foreign Ag depends on the
formation of an interface between two surfaces: the surface formed by
the TCR CDR loops and the surface formed by an Ag-presenting
molecule/Ag complex. Although MHC/peptide complexes have been
considered to be the primary ligands for the TCR CDR loop surface, it
is now clear that other ligands exist. CD1/lipid Ag complexes appear to
be recognized in much the same fashion as MHC/peptide complexes, in
that the TCR detects these two disparate complexes via the surface
formed by the CDR loops. The structural similarity between MHC and CD1
clearly is sufficient for the same TCR V and V elements to form
TCRs able to detect the chemically distinct universes of peptide and
lipid Ags. In the case of glycolipids, recognition of the Ag/CD1
complex may be similar to recognition of glycopeptides bound to MHC
class I molecules, in that in both cases the carbohydrate residues are
presumed to be the primary recognition elements that fit centrally into
the TCR binding surface (30, 31). Consequently, positive
and negative selection of T cells and the activation of T cells in the
periphery may include TCR-mediated interactions with both proteins and
lipids in the context of their respective Ag-presenting molecules, MHC
class I and II and CD1.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Millennium Pharmaceuticals, 75 Sidney Street, Cambridge, MA 02139.
3 Current address: Biogen, Inc., 14 Cambridge Center, Cambridge, MA 02142.
4 Current address: Department of Biocrystallography, Sincrotrone Trieste Elettra, SS 14 Km 163, 5 in Area Science Park, 34012 Basovizza (TS), Italy.
5 Current address: Department of Microbiology and Immunology, Albert Einstein College of Medicine, Room 416 Forchheimer Building, 1300 Morris Park Avenue, Bronx, NY 10461.
6 Current address: AstraZeneca Pharmaceuticals, 1800 Concord Pike, Wilmington, DE 19803.
7 Address correspondence and reprint requests to Dr. Michael B. Brenner, Brigham and Women’s Hospital, Smith Building, Room 514, 1 Jimmy Fund Way, Boston, MA 02115. E-mail address: mbrenner{at}rics.bwh.harvard.edu
8 Abbreviations used in this paper: CDR, complementarity-determining region; GMM, glucose monomycolate; SEC3, staphylococcal enterotoxin C3.
Received for publication October 11, 2001. Accepted for publication February 12, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
T cell receptor structure at 2.5 Angstrom and its orientation in the TCR-MHC complex. Science 274:209.+ T cells. Nature 372:691.[Medline]
T cell pool. J. Immunol. 162:366. TCRs. J. Immunol. 165:4494.3+ T cell receptors are consistent with an immunoglobulin-like structure for the receptor. J. Exp. Med. 177:119.-chain that interact with S. aureus toxin superantigens. Nature 346:471.[Medline]
-chain complexed with a superantigen. Nature 384:188.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D. C. Young and D. B. Moody T-cell recognition of glycolipids presented by CD1 proteins Glycobiology, July 1, 2006; 16(7): 103R - 112R. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Kjer-Nielsen, N. A. Borg, D. G. Pellicci, T. Beddoe, L. Kostenko, C. S. Clements, N. A. Williamson, M. J. Smyth, G. S. Besra, H. H. Reid, et al. A structural basis for selection and cross-species reactivity of the semi-invariant NKT cell receptor in CD1d/glycolipid recognition J. Exp. Med., March 20, 2006; 203(3): 661 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D. Gadola, M. Koch, J. Marles-Wright, N. M. Lissin, D. Shepherd, G. Matulis, K. Harlos, P. M. Villiger, D. I. Stuart, B. K. Jakobsen, et al. Structure and binding kinetics of three different human CD1d-{alpha}-galactosylceramide-specific T cell receptors J. Exp. Med., March 20, 2006; 203(3): 699 - 710. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Sieling, J. B. Torrelles, S. Stenger, W. Chung, A. E. Burdick, T. H. Rea, P. J. Brennan, J. T. Belisle, S. A. Porcelli, and R. L. Modlin The Human CD1-Restricted T Cell Repertoire Is Limited to Cross-Reactive Antigens: Implications for Host Responses against Immunologically Related Pathogens J. Immunol., March 1, 2005; 174(5): 2637 - 2644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gilleron, S. Stenger, Z. Mazorra, F. Wittke, S. Mariotti, G. Bohmer, J. Prandi, L. Mori, G. Puzo, and G. De Libero Diacylated Sulfoglycolipids Are Novel Mycobacterial Antigens Stimulating CD1-restricted T Cells during Infection with Mycobacterium tuberculosis J. Exp. Med., March 1, 2004; 199(5): 649 - 659. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Batuwangala, D. Shepherd, S. D. Gadola, K. J. C. Gibson, N. R. Zaccai, A. R. Fersht, G. S. Besra, V. Cerundolo, and E. Y. Jones The Crystal Structure of Human CD1b with a Bound Bacterial Glycolipid J. Immunol., February 15, 2004; 172(4): 2382 - 2388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Rauch, J. Gumperz, C. Robinson, M. Skold, C. Roy, D. C. Young, M. Lafleur, D. B. Moody, M. B. Brenner, C. E. Costello, et al. Structural Features of the Acyl Chain Determine Self-phospholipid Antigen Recognition by a CD1d-restricted Invariant NKT (iNKT) Cell J. Biol. Chem., November 28, 2003; 278(48): 47508 - 47515. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), wild-type (WT) DN1 TCR transfectants (
), and
mutated DN1 TCR transfectants R97A (
) and R98A (•) were cultured
with CD1+ monocytes and mycolic acids (A) or
SEC3 (B) for 24 h. IL-2 released into the
supernatants was then measured using the IL-2-dependent cell line HT-2.
C and D, Untransfected J. RT3-T3.5 cells
(
) and R98K (
) were cultured with
CD1+ monocytes and mycolic acids (C) or SEC3
(D), and IL-2 production was measured as in
A and B. Data represent the mean cpm
± SD of triplicate wells.
