| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
3 Domain1
*
Department of Biochemistry, University of Toronto, Toronto, Ontario, Canada;
Metabolism Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; and
R. W. Johnson Pharmaceutical Research Institute, San Diego, CA 92121
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Numerous studies have focused on proteins that participate in class I assembly. In mouse cells, newly synthesized class I H chains associate with calnexin (3, 4, 5), a membrane-bound chaperone of the ER, and this interaction facilitates H chain folding and promotes assembly with ß2m (6). Calnexin remains bound to the H chain following ß2m association and participates in retaining both the free H chain and H chain-ß2m assembly intermediates in the ER (7, 8, 9). Calreticulin, a soluble homologue of calnexin, has also been shown to associate with H chain-ß2m heterodimers but not free H chains in mouse cells (10, 11). The relative abundance of calnexin- vs calreticulin-associated class I heterodimers seems to be variable in different murine cells (9, 10, 11). Calnexin also binds to free H chains in human cells (4, 7, 12), enhancing H chain folding (13) and retaining this assembly intermediate in the ER (14). However, in contrast to the situation in mouse cells, calnexin largely dissociates upon H chain-ß2m assembly and is replaced by calreticulin (15, 16). Despite the demonstrated functions of calnexin, it is not essential in class I biogenesis, since class I assembly occurs normally in a calnexin-negative cell line (17, 18). However, it is likely that other ER chaperones, particularly calreticulin, can replace calnexin under these conditions. At present, the functions of calreticulin in class I biogenesis are poorly understood, although it probably participates in ER retention of peptide-deficient assembly intermediates (19). Furthermore, calnexin and calreticulin appear to play a major role in recruiting another protein, ERp57, into assembling class I complexes (see below).
In addition to chaperone interactions, H chain-ß2m heterodimers, but not free H chains, associate with TAP in both mouse and human cells (20, 21) forming large complexes that contain calnexin or calreticulin, H chain, ß2m, and TAP (9, 15). Stoichiometric analysis has revealed that as many as four H chain-ß2m heterodimers (and associated chaperones) may bind to a single TAP molecule (22). Very recently, a resident ER protein termed ERp57 has been detected in these complexes as well (23, 24, 25). ERp57 possesses both thiol oxidoreductase and cysteine protease activities, and it has been found associated with diverse, newly synthesized proteins in the ER (26, 27). It is likely that the association of ERp57 with class I-TAP complexes is mediated primarily through calnexin and calreticulin, since specific binding of ERp57 to both chaperones has been demonstrated (28) (M. Leach and D. B. Williams, unpublished observations), and the interaction of ERp57 with newly synthesized proteins (including class I) is prevented by treatment with castanospermine, an oligosaccharide processing inhibitor that abrogates the binding of calnexin and calreticulin to most glycoprotein substrates (23, 25, 26, 27). ERp57 has been proposed to participate in the oxidation or interchange of HC disulfide bonds, in the trimming of peptide ligands for class I within the ER, or in the reduction and degradation of misfolded or incompletely assembled class I molecules (23, 24, 25). The final step in class I assembly is the binding of high affinity peptides to the TAP-associated HC-ß2m heterodimer. This appears to be a critical event leading to dissociation of these large complexes and subsequent export of fully assembled class I proteins out of the ER (5, 9, 11, 20, 21).
It has been proposed that the TAP-class I interaction may enhance
peptide loading onto class I molecules, possibly by providing a high
local concentration of peptides (20, 21). However, there are
conflicting reports concerning the importance of the TAP-class I
interaction in promoting efficient assembly of class I molecules. It
has been reported that both membrane-bound and soluble forms of HLA-G
molecules contain similar sets of endogenous peptides despite the fact
that membrane-bound HLA-G, but not the soluble form, can be
coimmunoprecipitated with TAP (29). Furthermore, class I molecules
encoded by various HLA-B alleles appear not to associate with TAP as
assessed by coimmunoprecipitation and display no apparent defects in
their abilities to present antigenic peptides (30). Although these
findings argue against a role for the TAP-class I interaction in
enhancing peptide loading, it is uncertain whether the absence of
coimmunoprecipitable TAP-class I complexes in detergent lysates always
reflects a lack of TAP-class I interaction in the cell. The opposite
conclusion was reached in two studies using a mutant HLA-A2.1 molecule
containing a Thr to Lys mutation at residue 134 in the 
2 domain
(T134K) that cannot associate with TAP (31, 32). Although T134K was
fully capable of binding ß2m and peptide in vitro,
ß2m-associated forms of the molecule were unstable in
detergent lysates and were expressed at the cell surface at about 20%
the level of wild-type HLA-A2.1. These characteristics are similar to
those of peptide-deficient class I molecules produced in cells lacking
a functional TAP transporter (33, 34, 35). Furthermore, the T134K mutant
was substantially impaired in its ability to present peptides derived
from cytosolically expressed viral proteins to CTL (
0–25% relative
to HLA-A2.1). These findings appear to underscore the importance of the
TAP-class I interaction in peptide loading, but interpretation of the
data is complicated by the recent finding that the T134K mutant
molecule has also lost the ability to associate with the ER chaperone,
calreticulin (19). Consequently, it is not clear whether the impairment
in peptide loading of T134K is due solely to its inability to associate
with TAP.
Compelling support for the importance of the TAP-class I interaction in Ag presentation is based on studies of a mutant human lymphoblastoid cell line, 721.220. Surface levels of a variety of different class I molecules expressed in these cells are 20–25% of normal, and nascent class I molecules are labile in detergent lysates unless stabilized by exogenously added peptides (22, 36, 37). Furthermore, the ability to present cytosolically produced viral peptides to T cells is dramatically reduced (22). Although these cells have normal TAP function, the TAP-class I interaction is abolished due to the lack of tapasin, a 48-kDa type I membrane glycoprotein (15, 16, 22). Since complexes of tapasin can be isolated with class I in the absence of TAP or with TAP in the absence of class I, tapasin has been proposed to bridge the interaction between TAP and class I molecules (15). Transfection of tapasin cDNA into 721.220 cells restores TAP-class I association, normal cell surface class I expression, and presentation of viral Ag to CTL, thereby establishing the functional importance of tapasin in the TAP-class I interaction and providing support for the view that an association with TAP enhances peptide loading (22). Recently, this view has been called into question by a study in which the soluble, ER luminal domain of tapasin was expressed in 721.220 cells (38). Soluble tapasin retained the ability to associate with class I, but did not appear to interact with TAP. Remarkably, in the apparent absence of a TAP-class I interaction, soluble tapasin restored normal cell surface class I expression and the presentation of viral Ag to CTL. The authors concluded that tapasin itself is sufficient to promote peptide loading of class I molecules (38). It is noteworthy that soluble tapasin-class I complexes were unstable in detergent lysates and required chemical cross-linking for their visualization. Since a relatively small portion of TAP is predicted to reside within the ER lumen, the possibility remains that the inability to detect soluble tapasin-TAP complexes by cross-linking may reflect the lack of appropriately oriented functional groups rather than the absence of these complexes in vivo.
Although the presence of tapasin in human cells is well established, it
has been difficult to detect in murine TAP-class I complexes (9, 21).
In this study we demonstrate that tapasin is indeed associated with TAP
and class I molecules in mouse cells. In addition, we showed previously
that a Glu to Lys mutation at residue 222 in the 3 domain of
H-2Dd (designated Dd(E222K)) abrogates
association with TAP without affecting H chain-ß2m
assembly (9). We now show that Dd(E222K) fails to associate
with tapasin, thereby supporting a role for tapasin in bridging the
TAP-class I interaction and implicating residue 222 as a site of
contact with the tapasin/TAP complex. Furthermore, the lack of
association with tapasin and TAP impairs peptide loading and reduces
expression of Dd(E222K) at the cell surface. However, a
significant degree of peptide loading was retained, as evidenced by the
decay kinetics of surface Dd(E222K) molecules and by the
ability to accumulate target structures for CTL recognition. Thus,
although interaction with tapasin/TAP promotes efficient peptide
loading, it is not essential for acquisition of TAP-transported
peptides.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pSV2neo plasmids containing genomic constructs encoding the wild-type H-2Dd H chain or mutant Dd containing a Glu to Lys mutation at residue 222 (Dd(E222K)) were provided by Dr. Terry A. Potter (National Jewish Center for Immunology and Respiratory Diseases, Denver, CO) (39). The genomic constructs were transfected into murine thymoma BW5147 cells (H-2k haplotype) by electroporation. Transfected cells were selected in geneticin-containing medium (Life Technologies, Gaithersburg, MD) and then cloned by limiting dilution. Cells exhibiting a single symmetrical peak by flow cytometry using anti-Dd mAb 34-5-8S were judged to be clonal. Clones synthesizing similar levels of wild-type and E222K mutant Dd molecules were chosen for further study and were designated BW.Dd and BW.Dd(E222K), respectively. Bulk geneticin-resistant cells transfected with the pSV2neo plasmid were used as a negative control (BW.neo).
Cells, Abs, and other reagents
BW5147 transfectants were maintained in RPMI 1640 medium supplemented with 2 mM glutamine, 10% FCS, and antibiotics. L cells expressing wild-type or E222K mutant Dd molecules (gifts from Dr. Terry. A. Potter) were maintained in DMEM supplemented with 2 mM glutamine, 10% FCS, and antibiotics. The CD8-negative T cell hybridoma, B4.2.3, recognizes a peptide Ag derived from HIV-1 gp160 (residues 318–327; sequence RGPGRAFVTI) in an H-2Dd-restricted manner (40). B4.2.3 cells were maintained in RPMI 1640 medium supplemented with 2 mM glutamine, 10% FCS, and antibiotics.
The following mAb were used in this study: 34-5-8S and 34-4-20S, which
recognize ß2m-associated Dd molecules (41, 42); mAb 34-2-12S, which recognizes the 3 domain of Dd
molecules regardless of ß2m-association (41, 43); and mAb
16-3-1N which is specific for ß2m-associated
Kk molecules (44). Rabbit antisera specific for TAP1, TAP2,
and calnexin have been described previously (8, 21). Anti-tapasin
antiserum was raised in rabbits against a peptide corresponding to the
carboxyl-terminal 20 amino acids of murine tapasin (sequence:
SKEKATAASLTIPRNSKKSQ) (45). Rabbit anti-calreticulin antiserum was
purchased from Affinity Bioreagents (Golden, CO).
Carboxybenzyl-leucyl-leucyl-leucine vinyl sulfone (Z-L3VS) is a synthetic protease inhibitor that modifies the active site Thr of the catalytic ß subunits of the proteasome in a highly specific manner in vitro and in vivo (46). It was provided by Dr. Hidde Ploegh (Massachusetts Institute of Technology, Cambridge, MA). The following class I binding peptides were used in this study: tum-, a Dd-binding cellular peptide originally designated tum- 35B-F9 (sequence NGPPHSNNF) (47); p18-I10, a Dd-binding peptide corresponding to residues 318–327 of the HIV-1 envelope protein gp160 (sequence RGPGRAFVTI) (48, 49); and Flu NP Y367–374, a Db-binding peptide corresponding to residues 367–374 of the influenza A/PR/8/1934 nucleoprotein with an additional tyrosine at the amino terminus (sequence YSNENMETM) (50).
Metabolic radiolabeling and immunoisolation
Metabolic radiolabeling of L cell or BW5147 transfectants without or with chase incubation was performed as described previously (9, 21). For isolation of Dd molecules, 5 x 106 cells were lysed on ice for 30 min in 0.5 ml of Nonidet P-40 lysis buffer (1% Nonidet P-40 in PBS (pH 7.4), 10 mM iodoacetamide, 1% aprotinin, and 0.25 mM PMSF). Following centrifugation to remove nuclei and cell debris, the supernatant fraction was incubated on ice for 2 h with anti-Dd mAb, and then immune complexes were collected by shaking for 1 h with protein A-agarose. Immunoisolated proteins were eluted from protein A-agarose beads either directly or after endoglycosidase H (endo H) digestion and then were analyzed by SDS-PAGE (12.5% gel) followed by fluorography (8). For quantitation of bands, x-ray films were pre-exposed to white light to enhance the detection of weak bands and then exposed to the dried gel for durations chosen to ensure that the intensities of bands to be quantified were within the linear range of the film. Films were scanned using an EPSON 1000C scanner and were analyzed using National Institute of Health Image software. Background was subtracted by integrating a blank area of the film corresponding in size to that of the band to be quantified.
For detection of tapasin- or TAP-associated Dd molecules, cells were lysed in digitonin lysis buffer (0.5% digitonin in 10 mM HEPES (pH 7.2), 25 mM CaCl2, 10 mM iodoacetamide, 1% aprotinin, and 0.25 mM PMSF) and then subjected to sequential immunoisolation as described previously (9). Briefly, class I complexes with tapasin or TAP that were recovered in an initial round of immunoisolation with either anti-tapasin or anti-TAP2 antiserum were dissociated by heating at 40°C for 1 h in PBS, pH 7.4, containing 0.2% SDS. The solution was adjusted to contain 2% Nonidet P-40 and 5% skim milk powder and then was subjected to a second round of immunoisolation with mAb 34-2-12S. Immune complexes were washed only once with 0.5% Nonidet P-40, 10 mM Tris (pH 7.4), 150 mM NaCl, and 1 mM EDTA before elution and analysis by SDS-PAGE.
Flow cytometric analysis
To assess the levels of cell surface class I molecules, 1 x 106 cells were incubated with mouse anti-class I Abs (7 µg of mAb 34-5-8S for Dd or mAb 16-3-1N for Kk) for 30 min at 4°C in 0.25 ml of HBSS, 1% BSA, and 0.01% NaN3 (FACS buffer). After incubation, cells were washed once with FACS buffer and then were incubated with 5 µg of fluorescein-conjugated goat anti-mouse IgG (H+L chain specific; Jackson ImmunoResearch Laboratories, West Grove, PA) in 0.25 ml of FACS buffer for 30 min at 4°C. Cells were washed twice with wash buffer (HBSS and 0.01% NaN3) and then fixed in 1% paraformaldehyde in PBS, pH 7.4. Subsequent flow cytometric analysis was performed using an EPICS Elite flow cytometer (Coulter, Hialeah, FL).
For analysis of decay kinetics of cell surface class I molecules, cells were harvested at 4°C and washed with ice-cold protein-free RPMI 1640 medium. Cells were then resuspended at 106/ml in prewarmed (37°C) protein-free RPMI 1640 medium containing 10 µg/ml brefeldin A (Sigma, St. Louis, MO). A 1-ml aliquot was immediately removed and diluted into 2 ml of ice-cold FACS buffer. After centrifugation, the cells were processed at 4°C for flow cytometry as described above. The remaining cells were placed in a 37°C water bath in a CO2 incubator, and at various times 1-ml aliquots were removed and processed for flow cytometry.
T cell activation assay
The ability of cells to present exogenously loaded peptide Ag or HIV-1 gp160-derived Ag to T cells was evaluated by the amount of IL-2 produced by T cell hybridoma B4.2.3 cells upon recognition of its target structure (40, 51). Ltk- cells and L cell transfectants expressing wild-type Dd or Dd(E222K) were trypsinized, washed with PBS, and then either loaded with exogenous peptide or infected with recombinant vaccinia viruses. For peptide loading, cells were incubated with or without 1 µM p18-I10 peptide in RPMI 1640 medium for 4 h at 37°C. For viral infection, cells were incubated with either vSC8 (control vaccinia virus containing the lacZ gene) (52) or vPE16 (vaccinia virus containing the HIV-1 gp160 gene) (53) at 37°C in a CO2 incubator at either 5 or 20 pfu/cell. Excess viruses were washed out after 1 h and then incubated further for either 4 h (20 pfu/cell infection) or 16 h (5 pfu/cell infection). All cells were subsequently fixed by incubating in 5 µg/ml Psoralen (Sigma) for 10 min at room temperature followed by irradiation at 365 nm for 5 min. After washing, cells were plated in quadruplicate in a U-bottom 96-well plate at 5 x 104 cells/well and then were incubated overnight in a CO2 incubator with B4.2.3 cells (1 x 104 cells/well). The relative amount of IL-2 released into the supernatant fraction by B4.2.3 cells was measured by quantitating [3H]thymidine incorporation into the IL-2-dependent T cell line CT.EV during overnight culture (51).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Using transfected L cells, we showed previously that
Dd molecules possessing a Glu to Lys mutation at residue
222 fail to associate with the TAP peptide transporter (9). To help
clarify the basis for this loss of association, we examined the
interaction of wild-type Dd and Dd(E222K) with
tapasin. As shown in Fig. 1
A
(lane 1), when a digitonin lysate of L cells
expressing wild-type Dd was subjected to immunoisolation
with anti-tapasin antiserum, a major band was recovered with the
expected electrophoretic mobility of tapasin (48 kDa) as well as
additional bands with mobilities corresponding to TAP1 and TAP2
subunits, class I H chain, and ß2m. The identities of the
TAP subunits and the Dd H chain were confirmed by
dissociating the anti-tapasin immunoprecipitate in SDS and
recovering either TAP1 and TAP2 or Dd in a second round of
immunoprecipitation (Fig. 1A, lane 2, and Fig. 1B, lane 2, respectively). Faint bands
corresponding to calnexin and calreticulin were also identified by this
technique (Fig. 1A, lane 1, and data not shown).
These findings are consistent with results obtained using human cells
(15, 16, 22) and confirm that in murine cells tapasin is present in
complexes with TAP and class I molecules. Based on its mobility, the
band migrating slightly faster than calreticulin probably corresponds
to ERp57 (23, 24); the major band of about 100 kDa has not yet been
identified (Fig. 1A, asterisk).
|
B, lanes 2 and 6) or with an
isotype-matched irrelevant mAb (lanes 3 and
7). Although the resolution of this gel was somewhat poorer
than that shown in A, wild-type Dd H chains were
clearly recovered from the anti-tapasin immunoprecipitate, whereas
Dd(E222K) H chains could not be detected. Thus, as assessed
by coimmunoisolation, the Glu to Lys substitution in the 3 domain of
Dd prevents association with tapasin in addition to TAP.
HLA-A2.1 and H-2Ld molecules possessing point mutations at
H chain residues 134 and 227, respectively, do not form complexes with
TAP and, in addition, they have lost the ability to associate with
calreticulin (11, 31, 32). Consequently, it was important to determine
whether the failure of Dd(E222K) to associate with tapasin
and TAP might be an indirect consequence of altered binding to an ER
chaperone, specifically calnexin or calreticulin. To test this
possibility, radiolabeled lysates of L cells expressing Dd
or Dd(E222K) were subjected to a first round of
immunoprecipitation with anti-calnexin or anti-calreticulin
Abs, and then the immunoprecipitates were dissociated in SDS and
subjected to a second round of immunoprecipitation with
anti-Dd mAb. As shown in Fig. 1C, lane
3, wild-type Dd molecules could readily be recovered
from anti-calnexin immunoprecipitates, indicating interaction with
this chaperone. By contrast, no association with calreticulin was
observed (Fig. 1C, lane 5). Only upon prolonged
exposure (2–3 wk) could faint bands corresponding to wild-type
Dd molecules be detected from anti-calreticulin
immunoprecipitates (data not shown). We also examined chaperone binding
by immunoprecipitating Dd molecules with either
ß2m-dependent (34-5-8S) or ß2m-independent
(34-2-12S) mAbs and then immunoblotting the precipitates with
anti-calnexin or anti-calreticulin Abs. Again, calnexin was
readily detected in both Dd immunoprecipitates, but only
trace amounts of calreticulin were observed (data not shown). This
predominant interaction of wild-type Dd with calnexin, even
following ß2m association, differs from the results of
other studies in which murine Ld and Kb
molecules were shown to bind to both calnexin and calreticulin (10, 16). Presumably, the calreticulin band we detected in anti-tapasin
immunoprecipitates (Fig. 1A, lane 1) arises from
its association with endogenous k-haplotype class I molecules present
in L cells. Most importantly, mutant Dd(E222K) molecules
exhibited a pattern of chaperone interaction similar to that observed
for wild-type Dd. As shown in Fig. 1C,lanes 8 and 10, mutant molecules associated with
calnexin but not with calreticulin. However, trace interaction with
calreticulin could be detected upon prolonged exposure (data not
shown). Identical results were obtained by immunoprecipitating
Dd(E222K) and then immunoblotting with anti-calnexin or
anti-calreticulin Abs (data not shown). Therefore, the inability of
Dd(E222K) to bind tapasin and TAP cannot be attributed to
altered association with either chaperone.
To facilitate further investigation of the phenotype resulting from the lack of tapasin/TAP association, wild-type Dd or Dd(E222K) was expressed in BW5147 murine thymoma cells (H-2k), a nonadherent cell line that is more amenable to flow cytometric analysis and the production of large scale cultures. Cloned cells were chosen that synthesize similar levels of wild-type and mutant Dd (designated BW.Dd and BW.Dd(E222K)). Immunoprecipitation of cell lysates with anti-TAP2 or anti-tapasin antisera followed by SDS dissociation of immunoprecipitates and subsequent treatment with anti-Dd mAb confirmed that wild-type Dd, but not Dd(E222K), formed complexes with tapasin and TAP in these cells (data not shown).
We next examined to what extent the loss of tapasin/TAP association
affects the cell surface level of Dd molecules. It is well
established that in cells lacking a functional TAP transporter the
surface level of class I molecules is reduced 10- to 20-fold relative
to that in parental cells (33, 35). The decrease in surface expression
is due to the creation of "empty" class I molecules (H
chain-ß2m heterodimers devoid of peptide) that are
largely retained in the ER. Empty class I molecules are unstable, and
those that do reach the cell surface denature rapidly at 37°C unless
stabilized by exogenously added peptide or by culturing cells at low
temperature (54, 55). Distinct from the situation in TAP-deficient
cells, BW.Dd(E222K) should have normal TAP function, but
the Dd(E222K) molecules may be peptide deficient to some
degree due to their inability to associate with tapasin/TAP. Consistent
with this idea, flow cytometric analysis using the
conformation-sensitive anti-Dd mAb, 34-5-8S, revealed
that the surface level of Dd(E222K) is about 30%
that of wild-type Dd (Fig. 2). This reduced surface expression of
Dd(E222K) is not due to defect(s) in the Ag-processing
machinery that could have been introduced during transfection and
cloning because the levels of endogenous H-2Kk were
indistinguishable between BW.Dd and
BW.Dd(E222K) cells. It is also not a consequence of a
clonal peculiarity, because the bulk transfectants displayed a similar
pattern (data not shown).
|
We next sought to determine whether the low cell surface level of
Dd(E222K) is caused by inefficient peptide loading in the
ER. To address this question, cells were radiolabeled with
[35S]Met, and newly assembled Dd and
Dd(E222K) molecules were isolated with the
conformation-sensitive mAb, 34-5-8S, which recognizes only
ß2m-associated Dd. Even though Dd
and Dd(E222K) H chains were synthesized at similar rates
when analyzed with the ß2m-independent mAb 34-2-12S (Fig. 3A, left panel),
the amount of 34-5-8S-reactive Dd(E222K) was 44% that of
Dd (Fig. 3B, compare lanes 1 and
4). This was not due to an inherent defect in the ability of
Dd(E222K) H chains to associate with ß2m,
since the recovery of ß2m-associated
Dd(E222K) molecules could be enhanced by increasing the
protein concentration of cell lysates and by adding two
ß2m-dependent mAbs, 34-5-8S and 34-4-20S, at the time of
cell lysis (Fig. 3A, right panel). These
conditions favor the recovery of unstable H chain-ß2m
heterodimers (56).
|
B, compare lanes
4, 5, 6, and 7). In contrast,
wild-type Dd molecules were stable under the same
conditions (Fig. 3B, compare lanes 1 and
2). The results suggest that newly synthesized
Dd(E222K) molecules are either peptide deficient or
occupied with suboptimal peptides (peptides that deviate from the
canonical length or class I binding motif) that readily dissociate
(57). It is noteworthy that the addition of exogenous Dd
binding peptide increased the amounts of mAb 34-5-8S reactive molecules
for wild-type Dd as well as Dd(E222K) (Fig. 3B, compare lanes 1 vs 3 and
lanes 4 vs 6). This phenomenon has been observed
previously (34) and indicates that even in wild-type cell lysates there
are either free H chains and ß2m or unstable H
chain-ß2m heterodimers that can undergo peptide-driven
assembly or stabilization, respectively. Dd(E222K) molecules are transported inefficiently out of the ER and are unstable on the cell surface
In cells lacking a functional TAP transporter, empty class I
molecules are generally transported inefficiently from the ER and, upon
arrival at the cell surface, denature rapidly (55, 56, 58). To
determine whether Dd(E222K) molecules share these
characteristics, the kinetics of ER to Golgi transport of
Dd and Dd(E222K) molecules were examined in a
pulse-chase experiment using the ß2m-independent mAb
34-2-12S (Fig. 4). Processing of H chain
oligosaccharides to complex forms resistant to digestion by endo H was
used as a measure of the movement of Dd molecules from the
ER to the medial Golgi. Wild-type Dd was converted
quantitatively to endo H-resistant forms with a
t1/2 of approximately 45 min and remained stable
at the cell surface as expected for molecules possessing bound peptide
(Fig. 4, Dd). By contrast, only a small portion of
Dd(E222K) molecules was converted to endo H-resistant
forms, which were unstable and disappeared over time (Fig. 4, Dd(E222K), compare 2 h and 4 h time points).
These labile Dd(E222K) molecules could be stabilized by
adding a Dd-binding peptide and human ß2m to
the medium, suggesting that Dd(E222K) molecules reach the
cell surface either empty or with suboptimal peptides that readily
dissociate (Fig. 4, Dd(E222K) and
peptide/ß2m). However, conversion of
Dd(E222K) molecules to endo H-resistant forms was not
quantitative, suggesting that the majority of endo H-sensitive
Dd(E222K) molecules may be degraded intracellularly. This
view is supported by the finding that the rate of disappearance of endo
H-sensitive Dd(E222K) (t1/2
70 min) could be slowed by treating cells with Z-L3VS, a
specific inhibitor of proteasome-mediated degradation
(t1/2 155 min; data not shown).
Degradation is not related to the E222K mutation itself, since empty
wild-type Dd molecules produced in TAP-deficient LKD8c
cells exhibited similar behavior (data not shown). Therefore, in the
absence of an interaction with tapasin and TAP, the intracellular
transport of Dd(E222K) molecules resembles that of empty
class I molecules produced in TAP-deficient cells. Only a portion
(20%) reaches the cell surface where it exists transiently; the
majority appears to be retained in the ER and degraded. The retention
of the bulk of Dd(E222K) molecules in the absence of
interactions with tapasin and TAP is most likely mediated by calnexin,
since this chaperone associates with endo H-sensitive
Dd(E222K) molecules (Fig. 1C).
|
Stabilization of peptide-deficient cell surface class I molecules can be achieved by exogenous peptides of known class I binding ability or by culturing cells at lowered temperature (54, 59). These processes are highly dependent upon the presence of exogenous ß2m (provided as bovine ß2m in FCS or as purified human ß2m) (59), which drives the equilibrium of H chain interaction with ß2m toward H chain-ß2m heterodimers. To assess further the degree of peptide occupancy of cell surface Dd(E222K) molecules, the extent to which these molecules can be stabilized by peptide or low temperature was analyzed in comparison with wild-type Dd and with empty Dd molecules produced in TAP-deficient LKD8c cells (42).
In LKD8c cells, the basal level of surface Dd
molecules was very low when cells were cultured at 37°C in FCS alone
(Fig. 5, condition 9). Consistent with
previous observations (42), empty Dd molecules could be
stabilized and the surface level consequently increased by 33-fold
after incubating cells with exogenous peptide and ß2m at
37°C or by 14-fold following culture at 26°C in the presence of
ß2m (Fig. 5, compare condition 9 with conditions 10 and
12). Congruent results have been reported for empty Kb and
Db molecules produced in the TAP-deficient cell line
RMA-S (42, 54).
|
, compare condition 1 with conditions 2 and 4). The modest enhancement
reflects the fact that a substantial portion of wild-type
Dd molecules arrives at the cell surface containing stably
bound peptide.
Consistent with the data presented above (Fig. 2), the basal level of
Dd(E222K) surface expression was about 30% that of
wild-type Dd at 37°C (Fig. 5, conditions 1 and 5). This
basal expression could be enhanced in the presence of ß2m
by either peptide or low temperature by 5.6- and 3.8-fold, respectively
(Fig. 5, compare condition 5 with conditions 6 and 8). The fact that
these treatments enhanced the expression of Dd(E222K)
molecules to a much lesser extent than observed for empty class I
molecules in LKD8c cells suggests that a significant portion of
Dd(E222K) molecules reach the cell surface as peptide-bound
forms. However, compared with wild-type Dd, more
Dd(E222K) molecules appear to arrive at the cell surface
empty or with suboptimal peptides that readily dissociate.
Decay kinetics of surface Dd(E222K) molecules suggest the presence of peptide-containing species
Further evidence for the presence of peptide-containing
Dd(E222K) molecules was obtained by comparing the decay
kinetics of cell surface Dd(E222K) with those of wild-type
Dd and empty Dd molecules. BW.Dd,
BW.Dd(E222K), and trypsinized LKD8c cells were grown at
26°C in the presence of human ß2m to stabilize any
Dd molecules reaching the cell surface (conditions 4, 8,
and 12 in Fig. 5). Cells were then transferred to protein-free medium
containing brefeldin A to block further surface expression of newly
assembled Dd molecules. Following a shift to 37°C, the
levels of ß2m-associated Dd molecules on the
cell surface were assessed over a 2-h period by flow cytometry using
mAb 34-5-8S.
As shown in Fig. 6A, the decay
of Dd(E222K) was faster than that of wild-type
Dd but was substantially slower than the decay of empty
Dd molecules produced in TAP-deficient LKD8c cells. The
decay of Dd(E222K) could be slowed to a rate close to that
of wild-type Dd when BW.Dd(E222K) cells were
cultured in the presence of high affinity exogenous peptide and human
ß2m at 26°C (Fig. 6A). The intermediate
decay rate of Dd(E222K) molecules stabilized at 26°C is
most likely a combination of the decay rates of as many as three
populations: empty molecules, molecules containing peptides that are
indistinguishable in nature from those bound to wild-type
Dd, and molecules containing suboptimal peptides that
dissociate more rapidly than those bound to wild-type Dd.
Although it is difficult to quantify each population, we tested for the
presence of suboptimal peptides by comparing the decay kinetics of
wild-type Dd and Dd(E222K) molecules on cells
that were cultured at 37°C to eliminate empty molecules. As shown in
Fig. 6B, of the Dd(E222K) molecules that
remained after culture at 37°C (representing 27% of the amount
present after incubating at 26°C with ß2m; see Fig. 5),
a small, but highly reproducible, portion displayed faster decay
kinetics compared with wild-type Dd at the early phase of
the decay, suggesting the presence of suboptimal peptides. The
remainder of the Dd(E222K) molecules presumably contained
high affinity peptides and were relatively stable. In an effort to
visualize peptides bound to Dd(E222K) directly, we
immunoaffinity purified Dd and Dd(E222K)
molecules, eluted bound peptides, and analyzed them by reverse phase
HPLC. Unfortunately, the labile nature of Dd(E222K)
resulted in very poor recovery during affinity chromatography and
precluded a meaningful comparison of the peptide profiles (data not
shown).
|
Ag presentation by Dd(E222K)
Finally, we addressed the issue of whether the inability of Dd(E222K) to interact with tapasin and TAP affects its presentation of endogenously generated peptide Ags to T cells. The E222K mutation in the Dd H chain has been shown to abrogate CD8 binding as well as TAP association (39). Consequently, a CD8-negative and hence CD8-independent Dd-restricted T cell hybridoma, B4.2.3, was used in these experiments (40). The optimal peptide Ag for this T cell hybridoma is a 10-mer peptide termed p18-I10 (sequence: RGPGRAFVTI) corresponding to residues 318–327 of the HIV-1 envelope protein, gp160. Intracellular expression of gp160 protein was accomplished by infecting mouse L cells expressing either Dd or Dd(E222K) with a recombinant vaccinia virus containing the gp160 cDNA (53). L cells were used in these experiments because the BW5147 transfectants were not readily infectable with vaccinia virus. Consistent with the results obtained with the BW5147 transfectants, the cell surface level of Dd(E222K) was about 15% that of Dd in the L cells, although the two proteins were synthesized at similar rates (data not shown).
As shown in Fig. 7A, when
pulsed with exogenous p18-I10 peptide, uninfected L cells expressing
Dd(E222K) were able to stimulate the B4.2.3 hybridoma as
efficiently as L cells expressing wild-type Dd. Thus,
Dd(E222K) is fully capable of binding exogenously supplied
p18-I10 peptide and forming a TCR target structure that can be
recognized by the hybridoma. Indeed, the slightly higher level of
presentation by the mutant-expressing cells may be due to more
homogeneous loading with the p18-I10 peptide in the absence of as much
competition from endogenously loaded peptides. It was also important to
establish that under conditions of viral infection this peptide
determinant is generated in the cytosol and then transported into the
ER by TAP. The gp160 is a type I transmembrane protein that is
processed into an extracellular protein, gp120, and a transmembrane
protein, gp41, through proteolytic cleavage. Since the antigenic
sequence is located in the gp120 portion that passes transiently
through the ER, the possibility exists that the peptide determinant can
be generated by proteases within the ER and thereby bypass cytosolic
proteolysis and subsequent transport by TAP. If this is the case, any
advantage of the tapasin/TAP-class I association might be obscured.
Consequently, we tested whether the presentation of the gp160-derived
peptide to the B4.2.3 hybridoma is dependent on cytosolic proteasome
activity. As shown in Fig. 7B, the B4.2.3 hybridoma
specifically recognized Dd-expressing L cells that were
infected by vaccinia virus encoding gp160 but not by control vaccinia
virus containing the ß-galactosidase gene. Importantly, treatment of
cells during viral infection with the specific proteasome inhibitor,
Z-L3VS, completely blocked presentation of the
gp160-derived peptide to the B4.2.3 hybridoma, implicating the
proteasome in a crucial step of gp160 processing. Thus, the
gp160-derived peptide determinant presented to the B4.2.3 hybridoma
depends at some stage upon proteasomal cleavage in the cytosol and
presumably needs to be transported into the ER by TAP before binding to
Dd.
|
C,
right panel, the level of T cell activation by
Dd(E222K) was only 14% that achieved by wild-type
Dd when infection was conducted for 4 h (20 pfu/cell).
The difference was less pronounced if gp160 expression was allowed to
continue for a prolonged period of time; about a twofold difference was
observed after overnight infection (5 pfu/cell; Fig. 7C, left panel). These findings suggest that the
lack of tapasin and TAP interactions can be partially overcome by a
prolonged supply of antigenic peptide into the ER, leading to the
gradual accumulation of peptide-Dd(E222K) complexes at the
cell surface. Therefore, the tapasin/TAP-class I interaction appears
not to be essential for Ag presentation to T cells but, rather, it
enhances the efficiency of presentation. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Lys
substitution at residue 222 in the 3 domain. This mutant, that we
had previously determined not to bind to TAP (9), also fails to
associate with tapasin as judged by coimmunoisolation. The observation
that a single point mutation affects the binding of both tapasin and
TAP indicates that one protein relies on the other for association with
class I. Combined with previous studies showing that tapasin is
required for the association of class I and TAP (15, 36), this finding
supports the view that tapasin acts to bridge the interaction between
class I and TAP proteins. It is possible that the E222K mutation
permits some degree of H chain association with tapasin and TAP in vivo
that is not detectable under our conditions of detergent lysis and
immunoisolation. If such a weak association does exist in vivo it
appears not to be functionally relevant, since the phenotype of
Dd(E222K) molecules is similar to that of most class I
molecules expressed in the tapasin-negative cell line 721.220 (see
below).
It is unlikely that the E222K mutation causes major structural
alterations that indirectly affect the ability of the Dd
molecule to interact with tapasin. The mutation does not appear to
affect the association of the H chain with ß2m (a
prerequisite for tapasin binding) (15), since complexes of the E222K H
chain and murine ß2m can be isolated as efficiently as
those for wild-type Dd under conditions where dissociation
of peptide-deficient Dd(E222K) molecules is minimized.
Moreover, Dd(E222K) molecules are stabilized at the cell
surface at reduced temperature in the presence of exogenous human
ß2m. Similarly, the E222K mutation appears not to affect
peptide binding, since Dd(E222K) molecules can bind
peptides in detergent lysates or on the cell surface as evidenced by
the ability of exogenously added peptides (with ß2m) to
stabilize surface molecules at 37°C and to create T cell target
structures. Finally, the interaction of Dd(E222K) with
calnexin and calreticulin is not detectably altered relative to
wild-type Dd, rendering unlikely the possibility of
structural changes due to differences in molecular chaperone binding.
These findings suggest that residue 222 may be present within a segment
of the Dd H chain that contacts tapasin directly or that
mutation at this site causes relatively subtle structural changes that
affect other H chain-tapasin contact sites. In this context, it is
noteworthy that residue 222 comprises part of a conserved loop in the
3 domain (amino acids 222–229) that functions as a major binding
site for CD8 (39, 60). The crystal structure of the CD8-HLA-A2 complex
indicates that the two Ig domains of the CD8 homodimer bind to
the conserved 3 domain loop in the manner of an Ab-Ag complex (61).
Since tapasin has also been reported to contain two potential Ig
domains (22), it is conceivable that binding to the class I 3 domain
could occur in an analogous fashion.
Other studies have implicated the same region of the H chain 3
domain as being important for interaction with TAP. A mutant
Ld molecule possessing an AspLys point mutation at
residue 227 fails to coimmunoprecipitate with TAP (5). Furthermore, a
chimeric molecule in which residues 219–233 of the Db H
chain were replaced by residues 133–147 of the ß2 domain of the
class II I-Ad molecule also does not associate with TAP
(62). Interaction with tapasin was not examined in either case. Two
additional studies have pointed to the H chain 2 domain as important
for association with TAP (31, 32). Using transfected human C1R cells, a
point mutation at residue 134 in the HLA-A2.1 molecule (T134K) was
shown to result in loss of TAP association. However, this mutant as
well as the Ld(D227K) mutant described above also lost the
ability to bind to calreticulin; interactions with calnexin were
unaffected (11, 19). This raises the question of whether the lack of
TAP association is due to the loss of an important site of contact with
tapasin/TAP or to the loss of calreticulin interaction. Consistent with
the latter possibility is the finding that castanospermine treatment of
C1R cells, which prevents calreticulin and calnexin binding, inhibits
the formation of human class I-TAP complexes (15).
The functional consequence of the failure of Dd(E222K) to
associate with tapasin and TAP is a substantial impairment in the
loading of endogenously generated peptides. This is supported by
several observations that qualitatively resemble the phenotype of
peptide-deficient class I molecules expressed in cells lacking a
functional TAP transporter. First, the level of cell surface
Dd(E222K) is 15% (L cells) to 30% (BW5147 cells) that of
wild-type Dd, and expression can be up-regulated to
wild-type levels by the addition of exogenous peptides and
ß2m. Second, the majority of nascent
Dd(E222K) molecules are labile in detergent lysates unless
stabilized by the addition of exogenous Dd binding
peptides. Third, nascent Dd(E222K) molecules are
transported inefficiently to the cell surface, with the majority being
degraded intracellularly as endo H-sensitive forms. Finally, the
ability of Dd(E222K) molecules to present cytosol-derived
Ag to T cells is substantially impaired relative to that of wild-type
Dd. Our findings are largely in agreement with those
obtained with the human 721.220 cell line, in which class I molecules
fail to interact with TAP due to the absence of tapasin (15, 22, 36),
and also with studies on a chimeric Db molecule that does
not bind to TAP due to substitution of 3 domain residues 219–233
(62). The reduced peptide occupancy observed in all three instances
underscores the importance of the association between class I molecules
and tapasin/TAP for efficient loading of TAP-transported peptides.
It is noteworthy that significant residual peptide loading of the
Dd(E222K) mutant was observed in our study as evidenced by
its decay kinetics at the cell surface (Fig. 6A) and by its
ability to present cytosolically generated viral Ag to T cells (up to
50% relative to wild-type Dd after 16 h of
infection). By contrast, relatively little presentation of Ag to T
cells was observed following infection of tapasin-deficient 721.220
cells expressing HLA-A1, -B8, or -B*4402 molecules (22, 37) or cells
expressing either the Db (219–233) chimera (62) or the
A2.1(T134K) mutant (32). This may simply be a consequence of the
shorter infection times used in the latter studies (consistent with our
results after only 4 h of infection; Fig. 7C).
Alternatively, it might be argued that the E222K point mutation permits
a weak association with tapasin/TAP in vivo that cannot be detected by
coimmunoprecipitation. If this is the case, one would expect that a
similar spectrum of peptides would be acquired but with reduced
efficiency. However, comparison of the surface decay kinetics of
wild-type Dd with the population of Dd(E222K)
molecules that survives culture at 37°C revealed reproducible
differences (Fig. 6B). The data suggest that some
Dd(E222K) molecules possess suboptimal peptides that
dissociate more rapidly, thereby arguing against a low level of peptide
loading through weak interaction with tapasin and TAP. Rather, our
findings indicate that in the absence of association with tapasin/TAP,
peptides are still acquired but with reduced efficiency, consistent
with the fact that TAP continues to transport peptides into the lumen
of the ER. In agreement with this view, a very recent report has also
documented substantial presentation of Ag to T cells in
tapasin-deficient 721.220 cells following prolonged viral infection
(37).
The presence of suboptimal peptides in surface Dd(E222K) molecules suggests that association with tapasin/TAP provides, in addition to more efficient peptide loading, some type of peptide-editing function. While this manuscript was under review, McCluskey and co-workers reported evidence for acquisition of altered peptides in the absence of interactions with tapasin/TAP (37). They found that unlike other class I allotypes, HLA-B*2705 molecules could be expressed at the surface of 721.220 cells at a level similar to that observed in the tapasin-positive control. However, altered reactivity with a peptide-sensitive anti-HLA class I mAb was observed in the absence of tapasin, suggesting variation in the repertoire of peptides bound to HLA-B*2705. Perhaps tapasin/TAP retains class I molecules in the ER through cycles of binding of suboptimal peptides, releasing them only upon sensing conformational changes associated with the binding of optimal peptides. Alternatively, tapasin/TAP may actively facilitate dissociation of suboptimal peptides. Such a mechanism would be analogous to the function of HLA-DM in promoting acquisition of optimal peptides by MHC class II molecules (63). Independent evidence supporting an intracellular editing mechanism for class I molecules has been provided by the observation that a specific peptide-Kd complex dissociates more rapidly when retained in the ER than when expressed at the cell surface and that sequential peptide binding can occur in the ER (64). Furthermore, Lewis and Elliott have shown that the rapid export of unstable T134K HLA-A2.1 molecules (i.e., lacking optimal peptides) from the ER paradoxically requires a functional TAP peptide transporter (19). This has led to the hypothesis that the T134K mutant binds suboptimal TAP-transported peptides, which permit its escape from the ER quality control system. However, due to the mutation that interferes with binding to calreticulin and TAP, it fails to acquire optimal peptides and appears at the cell surface as an unstable molecule. By extension, wild-type molecules may go through similar cycles of peptide optimization.
The present study does not address the issue of how the class I-tapasin-TAP association enhances peptide loading. It has been thought that the proximity of class I to TAP promotes peptide delivery at least in part by providing a high local concentration of peptides. However, a recent study by Cresswell and co-workers, in which a soluble form of tapasin enhanced peptide loading of class I in the apparent absence of an interaction with TAP, raises the intriguing possibility that tapasin alone may be sufficient (38). Why, then, does the association with TAP exist? It may function only in regulating TAP levels as previously demonstrated (38). Alternatively, elevated peptide concentrations provided by the proximity of class I and TAP may be most relevant in cells where TAP is expressed at relatively low levels and peptide translocation rates are correspondingly low. In the experiments in which soluble tapasin was expressed in tapasin-deficient cells, TAP was present at high levels due to the transformation of these cells with EBV, and the increased rates of peptide translocation may have obscured any advantage of a physical proximity to TAP. It would be of interest to re-examine this issue in cells expressing lower levels of TAP.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, MA 02115.
3 Address correspondence and reprint requests to Dr. David B. Williams, Department of Biochemistry, Medical Sciences Building, University of Toronto, Toronto, Ontario, Canada, M5S 1A8. E-mail address:
4 Abbreviations used in this paper: H chain, class I heavy chain; ß2m, ß2-microglobulin; ER, endoplasmic reticulum; Z-L3VS, carboxybenzyl-leucyl-leucyl-leucine vinyl sulfone; pfu, plaque-forming units; Met, methionine; endo H, endoglycosidase H.
Received for publication May 6, 1998. Accepted for publication October 26, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2 domain of HLA-A2.1 define a functionally relevant interaction with TAP. Curr. Biol. 6:873.[Medline]
3 domain: CD8 and the T-cell receptor recognize the same class I molecule. Proc. Natl. Acad. Sci. USA 87:2137.3 domain of HLA-A2. Nature 345:41.[Medline]
and HLA-A2. Nature 387:630.[Medline]
3 domain with the transporter associated with antigen processing. J. Exp. Med. 187:865.This article has been cited by other articles:
|
|
 |
|
  Y. Zhang, G. Kozlov, C. L. Pocanschi, U. Brockmeier, B. S. Ireland, P. Maattanen, C. Howe, T. Elliott, K. Gehring, and D. B. Williams ERp57 Does Not Require Interactions with Calnexin and Calreticulin to Promote Assembly of Class I Histocompatibility Molecules, and It Enhances Peptide Loading Independently of Its Redox Activity J. Biol. Chem., April 10, 2009; 284(15): 10160 - 10173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. S. Ireland, U. Brockmeier, C. M. Howe, T. Elliott, and D. B. Williams Lectin-deficient Calreticulin Retains Full Functionality as a Chaperone for Class I Histocompatibility Molecules Mol. Biol. Cell, June 1, 2008; 19(6): 2413 - 2423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Lu, D. G. Kavanagh, and A. B. Hill Cellular and Molecular Requirements for Association of the Murine Cytomegalovirus Protein m4/gp34 with Major Histocompatibility Complex Class I Molecules. J. Virol., June 1, 2006; 80(12): 6048 - 6055. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Zhang, E. Baig, and D. B. Williams Functions of ERp57 in the Folding and Assembly of Major Histocompatibility Complex Class I Molecules J. Biol. Chem., May 26, 2006; 281(21): 14622 - 14631. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-E. Paquet, M. Cohen-Doyle, G. C. Shore, and D. B. Williams Bap29/31 Influences the Intracellular Traffic of MHC Class I Molecules J. Immunol., June 15, 2004; 172(12): 7548 - 7555. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. R. Morris, J. L. Petersen, S. E. Vargas, H. R. Turnquist, M. M. McIlhaney, S. D. Sanderson, J. T. Bruder, Y. Y. L. Yu, H.-G. Burgert, and J. C. Solheim The Amyloid Precursor-like Protein 2 and the Adenoviral E3/19K Protein Both Bind to a Conformational Site on H-2Kd and Regulate H-2Kd Expression J. Biol. Chem., April 4, 2003; 278(15): 12618 - 12623. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. O. Vertegaal, H. B. Kuiperij, A. Houweling, M. Verlaan, A. J. van der Eb, and A. Zantema Differential Expression of Tapasin and Immunoproteasome Subunits in Adenovirus Type 5- Versus Type 12-transformed Cells J. Biol. Chem., January 3, 2003; 278(1): 139 - 146. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.-E. Paquet and D. B. Williams Mutant MHC class I molecules define interactions between components of the peptide-loading complex Int. Immunol., April 1, 2002; 14(4): 347 - 358. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Tan, H. Kropshofer, O. Mandelboim, N. Bulbuc, G. J. Hammerling, and F. Momburg Recruitment of MHC Class I Molecules by Tapasin into the Transporter Associated with Antigen Processing-Associated Complex Is Essential for Optimal Peptide Loading J. Immunol., February 15, 2002; 168(4): 1950 - 1960. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. R. Turnquist, S. E. Vargas, A. J. Reber, M. M. McIlhaney, S. Li, P. Wang, S. D. Sanderson, B. Gubler, P. van Endert, and J. C. Solheim A Region of Tapasin That Affects Ld Binding and Assembly J. Immunol., October 15, 2001; 167(8): 4443 - 4449. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Harris, L. Lybarger, N. B. Myers, C. Hilbert, J. C. Solheim, T. H. Hansen, and Y. Y. L. Yu Interactions of HLA-B27 with the peptide loading complex as revealed by heavy chain mutations Int. Immunol., October 1, 2001; 13(10): 1275 - 1282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. E. Gewurz, R. Gaudet, D. Tortorella, E. W. Wang, H. L. Ploegh, and D. C. Wiley From the Cover: Antigen presentation subverted: Structure of the human cytomegalovirus protein US2 bound to the class I molecule HLA-A2 PNAS, June 5, 2001; 98(12): 6794 - 6799. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Harris, L. Lybarger, Y. Y. L. Yu, N. B. Myers, and T. H. Hansen Association of ERp57 with Mouse MHC Class I Molecules Is Tapasin Dependent and Mimics That of Calreticulin and not Calnexin J. Immunol., June 1, 2001; 166(11): 6686 - 6692. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. J. Hilton, A. M. Dahl, and K. L. Rock Anti-Peptide Antibody Blocks Peptide Binding to MHC Class I Molecules in the Endoplasmic Reticulum J. Immunol., March 15, 2001; 166(6): 3952 - 3956. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. L. Owen and L. R. Pease Thermal Stability of MHC Class I-{{beta}}2-Microglobulin Peptide Complexes in the Endoplasmic Reticulum Is Determined by the Peptide Occupancy of the Transporter Associated with Antigen Processing Complex J. Immunol., February 1, 2001; 166(3): 1740 - 1747. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Paulsson, P. O. Anderson, S. Chen, H.-O. Sjogren, H.-G. Ljunggren, P. Wang, and S. Li Assembly of tapasin-associated MHC class I in the absence of the transporter associated with antigen processing (TAP) Int. Immunol., January 1, 2001; 13(1): 23 - 29. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Myers, M. R. Harris, J. M. Connolly, L. Lybarger, Y. Y. L. Yu, and T. H. Hansen Kb, Kd, and Ld Molecules Share Common Tapasin Dependencies as Determined Using a Novel Epitope Tag J. Immunol., November 15, 2000; 165(10): 5656 - 5663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Barnden, A. W. Purcell, J. J. Gorman, and J. McCluskey Tapasin-Mediated Retention and Optimization of Peptide Ligands During the Assembly of Class I Molecules J. Immunol., July 1, 2000; 165(1): 322 - 330. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Li, K. M. Paulsson, S. Chen, H.-O. Sjogren, and P. Wang Tapasin Is Required for Efficient Peptide Binding to Transporter Associated with Antigen Processing J. Biol. Chem., January 21, 2000; 275(3): 1581 - 1586. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Peh, N. Laham, S. R. Burrows, Y. Zhu, and J. McCluskey Distinct Functions of Tapasin Revealed by Polymorphism in MHC Class I Peptide Loading J. Immunol., January 1, 2000; 164(1): 292 - 299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Lauvau, B. Gubler, H. Cohen, S. Daniel, S. Caillat-Zucman, and P. M. van Endert Tapasin Enhances Assembly of Transporters Associated with Antigen Processing-dependent and -independent Peptides with HLA-A2 and HLA-B27 Expressed in Insect Cells J. Biol. Chem., October 29, 1999; 274(44): 31349 - 31358. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Y. L. Yu, H. R. Turnquist, N. B. Myers, G. K. Balendiran, T. H. Hansen, and J. C. Solheim An Extensive Region of an MHC Class I {alpha}2 Domain Loop Influences Interaction with the Assembly Complex J. Immunol., October 15, 1999; 163(8): 4427 - 4433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. G. Danilczyk and D. B. Williams The Lectin Chaperone Calnexin Utilizes Polypeptide-based Interactions to Associate with Many of Its Substrates in Vivo J. Biol. Chem., June 29, 2001; 276(27): 25532 - 25540. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
