| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Division of Basic Immunology, Howard Hughes Medical Institute, National Jewish Medical and Research Center, Denver, CO 80206;
Department of Pathology, Center for Immunology, Washington University School of Medicine, St Louis, MO 63110; and
Departments of Immunology and of Biochemistry, Biophysics and Genetics, and
§
Departments of Immunology, Pharmacology and Medicine, University of Colorado Health Science Center, Denver, CO 80262
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As an alternative, we developed a method of attaching an
antigenic peptide to the N terminus of the MHC class II ß-chain via a
flexible peptide linker. The peptide fills the MHC binding groove
during biosynthesis, leading to production of a stable molecule fully
occupied by a single peptide. These molecules have been expressed by
ourselves and others in both soluble and membrane associated forms
5, 6, 7, 8, 9, 10 . For the vast majority of 
ßTCRs, these constructions are
recognized equivalently to MHC molecules bearing noncovalently bound
peptides, although an exception has been observed 11 .
Others have used a similar approach with MHC class I attaching the peptide via a flexible linker to the N terminus of the heavy chain 12 . Both membrane-bound and soluble forms of this construction have produced stable, functional MHC/peptide complexes. Most recently, a similar molecule with the peptide attached to the N terminus of ß2-microglobulin (ß2m)3 has been expressed as a cell surface molecule 13 .
In the present study, we used baculovirus to produce uniform, stable, soluble class I Dd molecules carrying peptides with a Dd-binding motif attached via a linker to the N terminus of ß2m. In the case of Dd carrying a peptide derived from HIV gp120, the secreted protein interacted specifically with a T cell hybridoma reactive to this combination.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A version of the gene encoding the heavy chain of the mouse
class I MHC molecule, Dd, was synthesized by PCR using a
cloned Dd cDNA as a template (kindly provided by Dr. David
Margulies, National Institutes of Health, Bethesda, MD). The PCR
fragment was cloned into a previously described baculovirus transfer
vector 5 behind the polyhedrin promoter. The gene was truncated at
the end of the 3 domain to remove sequence encoding the
transmembrane and cytoplasmic tail of the protein (Fig. 1
A). Another version of this
construct was made (Fig. 1B) including sequence at the 3'
end encoding a linker and a peptide tag capable of being biotinylated
by the Escherichia coli enzyme BirA 14, 15 . In both cases,
the gene was introduced into baculovirus using the BaculoGold system
(PharMingen, San Diego, CA) and a high-titered viral stock prepared as
previously described 5, 6 . The virus was used to infect High Five
insect cells (Invitrogen, San Diego, CA) as described in
Results.
|
DNA encoding full-length mouse ß2m (from C57BL/6)
was synthesized by PCR using a cDNA clone as template (kindly provided
by Dr. Terry Potter, National Jewish Medical and Research Center,
Denver, CO). The fragment was introduced into baculovirus using
the same transfer vector as above (Fig. 1
C). In two
additional constructs, the 5' end of the mouse ß2m gene
was modified to introduce sequence between that of the leader
and the ß2m N terminus encoding a flexible peptide linker
and either of two Dd-binding peptides (Fig. 1D).
One was a generic peptide, IGPARAAAL (pGPRL), and the other was the
well-characterized 16, 17 peptide from HIV gp120 corresponding to
amino acids 318–327 (pHIV). Both peptides contained the
Dd-binding motif, XGPXRXXX(X)L/I 18 . The genes were
introduced into baculovirus and high-titered viral stocks established
as above. These viruses were used with the Dd-producing
virus to coinfect High Five insect cells as described in
Results.
Monoclonal Abs
Two Dd-reactive mAbs were used in these experiments
18, 19 . mAb 34-2-12 is specific for the Dd heavy chain
3 domain and was used to capture Dd in an ELISA. mAb
34-5-8 is specific for Dd only when it contains bound
ß2m and a peptide in its binding groove. This Ab was used
for the detection of fully assembled Dd in an ELISA and for
immunoaffinity purification of fully assembled Dd. An
anti-mouse Cß mAb, H597 20 , was used in flow cytometric
experiments to detect surface ßTCR. CD8 was detected with the mAb,
53-6 21 . Bio-mAbs were produced using
sulfo-N-hydroxysuccinimide-LC-biotin (Pierce,
Rockford, IL).
ELISA for soluble assembled Dd
Immulon 1b microtiter plates (Dynex Technologies, Chantilly, VA) were coated with mAb 34-2-12. Various concentrations of culture supernatants containing an unknown amount of assembled Dd were incubated in the wells, and the bound Dd was detected with biotinylated mAb 34-5-8 and alkaline phosphatase-coupled Extravidin (Sigma, St. Louis, MO) using p-nitrophenylphosphate as substrate. A solution of purified Dd-pHIV of known concentration was used to establish a standard curve from which the concentrations of Dd in the unknown samples were estimated.
Cell lines
The CD8+ T cell hybridoma, B4.2.3 22 , was provided by Dr. David Margulies. This cell responds with IL-2 production to Dd plus pHIV. The Dd-bearing cell line, RMA/S-Dd, was kindly provided by Dr. David Raulet (University of California, Berkeley, CA). It is a version of the TAP-deficient cell line RMA/S that has been transfected with the gene for Dd heavy chain 23 . It was used to present free pHIV to B4.2.3. Another CD8+ T cell hybridoma, BD8-30, responsive to Kk as an alloantigen 24 , was used in control experiments. The CD4+ T cell hybridoma, KMAC-92, specific for Ek plus a peptide from moth cytochrome c corresponding to amino acids 88–103 (pMCC), has been previously described 25 .
Biotinylation of peptide tag and preparation of tetravalent Dd
Dd with pHIV or pGPRL covalently bound via ß2m was prepared carrying at its heavy chain C terminus the biotinylation peptide tag described above. The proteins were affinity-purified and subjected to size exclusion chromatography as described in Results. Aliquots were enzymatically biotinylated with BirA and tetravalent complexes of the proteins with phycoerythrin-streptavidin (PESA) were prepared, purified, and stored as previously described 15 . Similarly, a tetravalent version of biotinylated soluble Ek covalently bound to pMCC was prepared.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Insect cells were coinfected at high multiplicity with various
ratios of baculovirus encoding Dd heavy chain and either
pGPRL-ß2m, pHIV-ß2m, or ß2m
without a covalent peptide. After 5 days, culture supernatants were
assayed for assembled Dd molecules using an ELISA. An
3-specific mAb was used to capture the soluble Dd, and a
mAb specific for fully assembled, peptide-containing
Dd was used for detection. The results are shown in Fig. 2. As predicted by the specificities of
the mAbs used in the ELISA, no product was detected when the insect
cells were infected with either Dd heavy chain or one of
the ß2m-producing viruses alone (Fig. 2, A-C).
Only trace amounts of assembled Dd were detected after
coinfection with viruses producing heavy chain and free
ß2m (Fig. 2A). This result indicated that, for
the most part, the appropriate peptides for binding stably to
Dd were missing from the insect cells and the medium. High
levels of assembled Dd were detected after coinfection with
optimal ratios of viruses producing Dd heavy chain and
ß2m covalently attached to either pGPRL or pHIV (Fig. 2, B and C). These results indicated that the
soluble Dd heavy chain associated properly with the peptide
linked to ß2m and that the attached peptide correctly
occupied the peptide binding groove.
|
A liter of culture supernatant was produced from High Five insect
cells coinfected with the optimal ratio of Dd heavy chain
and pHIV-ß2m-producing viruses. The secreted
Dd-pHIV protein was immunoaffinity-purified on a mAb 34-5-8
column and eluted with a buffer (pH 10.8). The eluate was neutralized,
concentrated, and subjected to size exclusion chromatography on
Superdex-200 (Pharmacia, Piscataway, NJ) (Fig. 3A). Fractions containing the
major peak were pooled and concentrated. The final yield was 
4 mg.
An aliquot of the concentrate was analyzed by SDS-PAGE (Fig. 3B) and isoelectric focusing (Fig. 3C) using the
Phast system (Pharmacia). The SDS-PAGE gels showed that the
preparation contained highly purified Dd, although some
heterogeneity was evident in the heavy chain, perhaps due to
carbohydrate heterogeneity. Significantly, with SDS-PAGE, the nearly
uniform band corresponding to pHIV-ß2m showed
an apparent m.w. of 14 kDa, reflecting the uniform increase in size of
ß2m predicted by the addition of the linker and
peptide. The preparation had an average isoelectric point of 6.7 with
only a small amount of charge heterogeneity.
|
ßTCR
In the case of peptides covalently attached to MHC class II
molecules, the extension of the peptide from the peptide binding groove
allows the attachment of the covalent linker in such a way as to
minimize the possible interference with T cell recognition of the
peptide/MHC complex 26 . Since the C terminus of the peptides bound to
MHC class I is normally intimately involved in the peptide binding
groove, the potential for the covalent linker to interfere with T cell
recognition is higher. Therefore, we tested the ability of the
Dd-pHIV construct to be recognized by a T cell hybridoma,
B4.2.3, originally identified as specific for the free HIV peptide plus
Dd 17, 22 . Biotinylated versions of Dd-pHIV
and Dd-pGPRL were immobilized in tissue culture wells, and
their ability to stimulate IL-2 production by B4.2.3 was tested. The
results (Fig. 4) showed that
Dd-pHIV, but not Dd-pGPRL, stimulated IL-2
production by B4.2.3 in a dose-dependent manner.
|
. All three hybridomas
bound the anti-Cß mAb, indicating the level of their surface
ßTCR expression (Fig. 5, A-C). As expected, the two
class I MHC-specific hybridomas bound anti-CD8, but the class II
MHC-specific hybridoma did not (Fig. 5, A-C). The
PESA-Dd-pHIV reagent bound to B4.2.3 (Fig. 5D),
but not to BD8-30 or KMAC-92 (Fig. 5, E and F).
The PESA-Dd-pGPRL reagent did not bind to any of the
hybridomas (Fig. 5, D-F). Finally, as expected,
PESA-I-Ek-pMCC reagent bound specifically to KMAC-92 (Fig. 5F). These results show the dependence of the binding on the
correct peptide bound to Dd. They also show that the
tetravalent Dd did not bind to CD8 in the absence of the
specific ßTCR, since neither class I reagent bound to BD8-30,
despite its very high surface expression of CD8.
|
, G and H). In a previous study 15 , we showed
that the failure to reach saturation leads to a direct relationship
between the extent of multivalent MHC-peptide binding to T cell
hybridomas and the affinity of the ßTCR for the monovalent ligand
measured in vitro. We tested the idea that this principal applies as
well to the binding of PESA-Dd-pHIV to B4.2.3. In Fig. 6, the open symbols show the previously
reported 15 linear relation between tetrameric class II MHC-peptide
binding to five different T cell hybridomas (including KMAC-92) and the
dissociation constants of their ßTCRs established by surface
plasmon resonance. We calculated the normalized binding of tetravalent
Dd-pHIV to B4.2.3 and Ek-pMCC to KMAC-92 from
the data in Fig. 5 (A, C, D, and
F) and plotted these values vs the previously measured
dissociation constants of the hybridoma ßTCRs for their ligands:
30 µM for KMAC-92 27 and 2–8 µM for B4.2.3 28 . The current
data for KMAC-92 (closed triangle) agree very well with the previously
determined value (open triangle). The data for B4.2.3 show that this
same standard curve predicts as well the relation between the binding
of tetravalent class I Dd-pHIV to this hybridoma and the
known affinity of its ßTCR. These data strengthen our conclusion
that the linker attaching pHIV to ß2m did not alter the
affinity of the B4.2.3 ßTCR for its ligand.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ßTCRs either in affinity measurements
27, 28, 29, 30, 31, 32, 33, 34 or in directly visualizing Ag-specific T cells 15, 35, 36, 37, 38, 39 .
In the latter case, multimerization of the MHC molecules overcomes the
naturally low affinity of ßTCRs for MHC ligands, producing a
higher avidity via multipoint binding. In both of these types of
studies, the quality of the reagent depends on the complete occupancy
of the MHC peptide binding groove with a single peptide. While this has
been achieved for both class I and class II MHC by loading "empty"
MHC molecules in the presence of high concentrations of the peptide,
the inherent instability of MHC molecules in the absence of a peptide
can reduce yield and, for certain MHC alleles and isotypes, result in
denaturation without significant peptide loading. One alternate approach has been the covalent attachment of the peptide to the MHC molecule via a linker to the peptide C terminus 5, 6 . The structure of the peptide binding groove of class II MHC allows longer peptides to exit the groove at both the N-terminal and C-terminal ends. Therefore, the attachment of a linker to the C-terminal end was predicted not to interfere with peptide binding or T cell recognition of the complex. The prediction was borne out by the crystal structures of numerous class II MHC molecules with covalently attached peptides 9, 10, 26 .
For class I MHC, the C terminus of the peptide is intimately involved
in MHC binding, and the groove is closed at both ends. It was not at
first apparent how a linker could be attached to the peptide C terminus
without disrupting peptide binding and/or interfering with T cell
recognition. However, the many crystal structures of the MHC class
I/peptide molecules offer an explanation for the success of these
covalent constructs. In these structures, the peptide C-terminal
carboxylate is firmly anchored in the groove via its carbonyl group,
but its hydroxyl group is exposed on the surface for possible extension
of the peptide. Such a case has been seen with an unusual peptide in
which this exposed hydroxyl forms a peptide bond with a glycine
extending the peptide by one residue that is fully solvent-exposed
40 . Furthermore, the recent crystal structures of soluble TCRs bound
to MHC class I ligands 41, 42 suggest that, depending on the exact
length and position of the Vß CDR1 loop, room exists for the
peptide linker attached to the peptide C terminus to exit the interface
without disrupting ßTCR binding.
The method we describe here of attaching the peptide to the N terminus
of ß2m via a flexible linker results in a high yield of
fully assembled, soluble MHC class I occupied by a single peptide and
suitable for ßTCR interaction studies. We chose to attach the
peptide to the N terminus of ß2m rather than MHC heavy
chain because the crystal structures of class I predict a more direct
path for the linker from the peptide C terminus to the
ß2m N terminus. We chose a linker length of 13 amino
acids, again based on crystal structures of MHC class I, predicting
that the path from the peptide C terminus to the ß2m N
terminus could be comfortably covered by a flexible linker of this
length without interfering with ßTCR binding.
Our experiments with Dd-pHIV support this rationale,
since this covalent ligand interacted specifically with an ßTCR of
a T cell hybridoma raised to Dd plus a free HIV peptide.
This interaction was demonstrated both by direct binding of a
tetravalent Dd-pHIV to the hybridoma and by the stimulation
by immobilized Dd-pHIV of the response of the hybridoma.
After we completed these studies, another group reported success with attaching peptides to the N terminus of ß2m via a linker 13 . In this case, the presentation of peptide to T cells by cells expressing the natural membrane-anchored form of either the Db or Kd heavy chain was studied. In one case, the gene for the peptide-ß2m was coexpressed with the heavy chain. In the other, the peptide-ß2m protein produced in bacteria was incubated with the heavy chain-expressing cells. In both cases, the cells became sensitive to lysis by peptide-specific T cells. While these authors did not establish directly the percent of class I occupied by the peptide-ß2m, they did offer evidence that the T cells were recognizing class I in which the peptide to ß2m covalent linkage was intact. Thus, in at least three cases, this type of linkage does not appear to interfere with T cell recognition.
In a previous study 15 , we established the direct relationship
between the extent of tetrameric class II MHC-peptide binding to a T
cell hybridoma and the affinity of the hybridoma ßTCR for
monomeric MHC-peptide. Our data presented here suggest that this
principle may apply as well to class I MHC-peptide tetramers, although
additional examples will be needed to test the hypothesis rigorously.
Recent work has established the usefulness of tetravalent reagents in
detecting Ag-specific cells 15, 35, 36, 37, 38, 39 . Our findings offer the
prospect of being able to both enumerate class I and class II
restricted Ag-specific T cells and to estimate easily the affinity of
their receptors for peptide/MHC ligands. Such tools may offer insights
into the problems of determinant selection in T cell immune responses
and the escape of autoreactive T cells from elimination during the
establishment of self-tolerance.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. John Kappler, Howard Hughes Medical Institute, National Jewish Medical and Research Center, 1400 Jackson Street, Denver, CO 80206. E-mail address:
3 Abbreviations used in this paper: ß2m, ß2-microglobulin; PESA, phycoerythrin-streptavidin; pGPRL, peptide IGPARAAAL; pHIV, HIV gp120 amino acids 318–327; pMCC, moth cytochrome c amino acids 88–103.
Received for publication June 25, 1998. Accepted for publication November 20, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß heterodimers in the absence of antigenic peptide. Cell 68:465.[Medline]
ß T cell receptors. J. Immunol. 142:2736.[Abstract]
ß T cell receptor interactions with syngeneic and allogeneic ligands: affinity measurements and crystallization. Proc. Natl. Acad. Sci. USA 94:13838.ß T-cell receptor and its superantigen and class II-MHC/peptide ligands. Mol. Immunol. 34:493.[Medline]
chain on major histocompatibility complex (MHC) class II+ peptide recognition. J. Exp. Med. 185:1919.This article has been cited by other articles:
|
|
 |
|
  A. Margalit, H. M. Sheikhet, Y. Carmi, D. Berko, E. Tzehoval, L. Eisenbach, and G. Gross Induction of Antitumor Immunity by CTL Epitopes Genetically Linked to Membrane-Anchored {beta}2-Microglobulin J. Immunol., January 1, 2006; 176(1): 217 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, A. Rubtsov, R. Heiser, J. White, F. Crawford, P. Marrack, and J. W. Kappler Using a baculovirus display library to identify MHC class I mimotopes PNAS, February 15, 2005; 102(7): 2476 - 2481. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Berko, Y. Carmi, G. Cafri, S. Ben-Zaken, H. M. Sheikhet, E. Tzehoval, L. Eisenbach, A. Margalit, and G. Gross Membrane-Anchored {beta}2-Microglobulin Stabilizes a Highly Receptive State of MHC Class I Molecules J. Immunol., February 15, 2005; 174(4): 2116 - 2123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Margalit, S. Fishman, D. Berko, J. Engberg, and G. Gross Chimeric {beta}2 microglobulin/CD3{zeta} polypeptides expressed in T cells convert MHC class I peptide ligands into T cell activation receptors: a potential tool for specific targeting of pathogenic CD8+ T cells Int. Immunol., November 1, 2003; 15(11): 1379 - 1387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Lybarger, Y. Y. L. Yu, M. J. Miley, D. H. Fremont, N. Myers, T. Primeau, S. M. Truscott, J. M. Connolly, and T. H. Hansen Enhanced Immune Presentation of a Single-chain Major Histocompatibility Complex Class I Molecule Engineered to Optimize Linkage of a C-terminally Extended Peptide J. Biol. Chem., July 11, 2003; 278(29): 27105 - 27111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kawana-Tachikawa, M. Tomizawa, J.-i. Nunoya, T. Shioda, A. Kato, E. E. Nakayama, T. Nakamura, Y. Nagai, and A. Iwamoto An Efficient and Versatile Mammalian Viral Vector System for Major Histocompatibility Complex Class I/Peptide Complexes J. Virol., October 25, 2002; 76(23): 11982 - 11988. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Lev, H. Novak, D. Segal, and Y. Reiter Recruitment of CTL Activity by Tumor-Specific Antibody-Mediated Targeting of Single-Chain Class I MHC-Peptide Complexes J. Immunol., September 15, 2002; 169(6): 2988 - 2996. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Y. L. Yu, N. Netuschil, L. Lybarger, J. M. Connolly, and T. H. Hansen Cutting Edge: Single-Chain Trimers of MHC Class I Molecules Form Stable Structures That Potently Stimulate Antigen-Specific T Cells and B Cells J. Immunol., April 1, 2002; 168(7): 3145 - 3149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fasso, N. Anandasabapathy, F. Crawford, J. Kappler, C. G. Fathman, and W. M. Ridgway T Cell Receptor (TCR)-mediated Repertoire Selection and Loss of TCR V{beta} Diversity during the Initiation of a CD4+ T Cell Response In Vivo J. Exp. Med., December 11, 2000; 192(12): 1719 - 1730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Kedl, W. A. Rees, D. A. Hildeman, B. Schaefer, T. Mitchell, J. Kappler, and P. Marrack T Cells Compete for Access to Antigen-bearing Antigen-presenting Cells J. Exp. Med., October 9, 2000; 192(8): 1105 - 1114. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
). The plates were washed and
105 B4.2.3 hybridoma cells were added. As a control,
105 B4.2.3 were added to wells containing 105
RMA/S-Dd APCs and 10 µg/ml of pHIV (solid bar). After
24 h, the culture supernatants were assayed for IL-2.
) was
normalized to the binding of anti-Cß using the data shown in Fig. 5
), KMAC-92 (
), KMAC-126 (
) and DO-11.10 (
).