| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Genetics, Washington University School of Medicine, St. Louis, MO 63110;
Committee on Immunology and Pathology, Gwen Knapp Center for Lupus and Immunology Research, University of Chicago, Chicago, IL 60637; and
Department of Microbiology and Immunology, Howard Hughes Medical Institute, Vanderbilt University School of Medicine, Nashville, TN 37232
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
2-microglobulin
(2m),3
and antigenic peptide, and represents the end product of a complex
biosynthetic pathway. This pathway is critical for the expression of
class I molecules and has been researched extensively. Many of the key
steps in class I maturation occur within the endoplasmic reticulum (ER)
and invoke the participation of several ER-resident chaperone molecules
(1, 2, 3). These chaperones include TAP-1 and TAP-2, tapasin,
calnexin, calreticulin, and ERp57. Together, these molecules assist in
the formation of H chain/2m/peptide trimers
that, once stably assembled, transit to the cell surface. In accordance
with the concerted role of these chaperones in class I biosynthesis,
they are all present in a molecular complex that also contains H chain
and 2m, referred to in this work as the
peptide-loading complex. Of the members of this loading complex, only
TAP-1/2 and tapasin appear to be involved solely with class I
biosynthesis. The TAP-1/2 heterodimer forms a peptide pump that
translocates antigenic peptides from the cytosol to the ER lumen
(1, 2). Tapasin clearly facilitates expression of most
class I molecules (4, 5, 6, 7), but its precise function(s) in
this process has not been fully elucidated. Multiple functions have been ascribed to tapasin, including ER retention of peptide-empty class I, facilitation of peptide binding, and selection of high-affinity peptides (peptide editing) (8). In addition, tapasin bridges the H chain with TAP-1/2 (9) and stabilizes the TAP heterodimer (10). Studies using soluble tapasin that does not promote TAP association (10) suggested that tapasin may directly facilitate H chain folding (10, 11), a notion supported by studies of class I folding in cell lysates (12). Furthermore, a role for tapasin in the retention of empty class I molecules within the ER has also been proposed. Reconstitution of insect cells with Kb alone resulted in the surface expression of empty molecules, while coexpression of tapasin resulted in the efficient ER retention of Kb (13). In a separate study, Kb expression was analyzed in tapasin-deficient human .220 cells, and it was found to exit from the ER with increased kinetics in the absence of tapasin (14). Similarly, Grandea et al. (6) demonstrated that peptide-accessible Kb and Db molecules escaped the ER in cells from tapasin-deficient mice, but were reliably retained in tapasin-positive cells. In these studies of class I expression in mammalian cells, it is difficult to determine with certainty whether the molecules that escape the ER are truly peptide empty, or if they instead leave the ER in a folded conformation, but contain lower affinity ligands. It should be noted that the disparate quality of ligands presented at the cell surface of tapasin-positive vs tapasin-negative cells could result from any of the proposed functions of tapasin (TAP stabilization, ER retention, H chain folding, and peptide editing) or any combination of these functions. Indeed, it has proven difficult to determine which of these interrelated functions are primary vs secondary manifestations of tapasin’s interaction with class I.
In addition to the uncertainty regarding the precise functions of tapasin, significant differences have been noted in the relative levels of TAP/tapasin association between different class I molecules and allelic gene products (15). Although these differences in peptide-loading complex association may reflect structural differences between class I molecules, they may also reflect the fact that class I molecules have to compete for TAP/tapasin docking sites (16). Furthermore, there is clear evidence that the size of the peptide pool capable of binding to a given class I molecule is inversely correlated with the steady-state interaction of that molecule with the loading complex (17, 18). In addition to these apparent differences in the levels of class I association with TAP/tapasin, various class I molecules have been reported to differ in their dependency on tapasin for peptide binding. For example, certain HLA molecules were found to be relatively tapasin independent in their surface expression, whereas others were found to be relatively tapasin dependent (19). One confounding factor in determining the differences in the relative tapasin dependencies of various class I molecules is the differences between the respective peptide pools that are present for different molecules. Furthermore, as mentioned above, tapasin dependency could reflect either a direct role for tapasin in H chain folding, a requirement for tapasin-mediated ER retention of peptide-empty forms to achieve binding of high-affinity ligands, TAP stabilization, or a combination of these activities. In fact, different class I molecules could vary in their relative dependence on the different functions of tapasin.
To address some of these outstanding questions, we sought an experimental system that would permit a direct test of the putative tapasin functions independent of one another. In this regard, we considered it relevant to study the biosynthesis of the MHC class Ib molecule, H2-M3. M3 is a MHC-encoded (H2-M3) protein with the capacity to present relatively short peptide Ags (five to seven residues) that possess an N-terminal formyl moiety (20). Since the N termini of prokaryotic proteins initiate with formylated methionine residues, M3 molecules can function as restriction elements for CD8 T cell responses against intracellular bacteria. In contrast to class Ia molecules, M3 is expressed at very low levels on the surface of normal cells (21, 22) presumably due to the rarity of endogenous formylated methionine-containing peptides that could only arise from mitochondrial proteins. Despite the low level of constitutive surface expression of M3, Chiu et al. (22) demonstrated recently that there is an extensive pool of peptide-receptive M3 molecules in the ER that can be brought to the cell surface by culturing cells with exogenous M3 peptide ligands. This induction was largely TAP dependent, as are some (23, 24), but not all (25) T cell responses to M3. These properties of M3 present a unique opportunity to probe both 1) how a highly restricted pool of peptide ligands affects the steady-state association of a class I molecule with the peptide-loading complex, and 2) the role of tapasin in ER retention of an extensive pool of open class I molecules.
In this study, we used a novel epitope tag strategy that permitted a direct comparison of the chaperone interactions of M3 vs class Ia molecules, using the same anti-H chain mAb. We present three lines of evidence that M3 assembly is regulated differently than assembly of class Ia molecules. First, intrinsic properties of M3, independent of its restricted pool of endogenous ligands, determine its steady-state interactions with the mouse or human peptide-loading complex. Second, a novel factor absent in certain human cell lines is required for peptide-induced folding of M3, but not class Ia molecules. Third, peptide-empty forms of M3 are not present at the cell surface in tapasin-deficient cells, in contrast to class Ia molecules. However, tapasin can facilitate peptide binding in the ER by a mechanism independent of TAP stabilization and ER retention.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
To introduce the 64-3-7 epitope (which is specific for open forms of Ld and Lq), a single substitution (R48Q) was required to convert the sequence of M3 surrounding the epitope to the sequence found in Ld (26). Site-directed mutagenesis of the wild-type M3 cDNA was performed using the Quik Change mutagenesis kit (Stratagene, La Jolla, CA) essentially as described (26). The following primers were used to introduce the R48Q mutation: sense primer, CCG AGG ATG GAA CCT CAA GCA CCA TGG ATG GAG AAG GAA AGA CC; antisense primer, GGT CTT TCC TTC TCC ATC CAT GGT GCT TGA GGT TCC ATC CTC GG. DNA sequence analysis was used to confirm the presence of the mutation on the resulting epitope-tagged M3 cDNA (etM3). The etM3 cDNA was inserted into the mammalian expression vector pIRESneo (CLONTECH Laboratories, Palo Alto, CA) for all transfections. To construct the etM3/yellow fluorescent protein (YFP) chimera, the respective cDNAs were PCR amplified and the EYFP coding sequence (from pEYFP-C1; CLONTECH Laboratories) was appended to the cytoplasmic tail of etM3. The last residue in M3 was joined to the first residue of YFP by an ala-ser spacer. DNA sequence analysis verified the correct sequence of the chimeric construct. This construct, as well as the full-length murine tapasin cDNA and the soluble tapasin cDNA were expressed from pIRESneo. A soluble murine tapasin construct was generated that includes residues 1–384 and, thus, lacks the transmembrane segment and cytoplasmic tail (10).
Cell lines and transfections
Mouse L cell transfectants of Ld (L-Ld) have been described elsewhere (27). An etM3-expressing DAP-3 fibroblast line was generated by transfection of the parental line with the etM3 cDNA using Lipofectin (Life Technologies, Gaithersburg, MD). Stable expressors were selected with 0.6 mg/ml geneticin (Life Technologies), identified by intracellular staining/flow cytometry (see below), and cloned by limiting dilution. etM3 transfectants of the human cervical carcinoma line, HeLa, were also generated using Lipofectin, followed by selection in 0.6 mg/ml geneticin. The human B lymphoblastoid 721.221 cell line (28) was a gift from T. Spies (University of Washington, Seattle, WA). Transfection of these cells with etM3 was accomplished via electroporation using the Gene Pulser II System (Bio-Rad, Hercules, CA). Briefly, 8 x 106 cells (at 107 cells/ml in PBS) were mixed with 10 µg plasmid DNA and pulsed with 400 V at 950 mF. Stably expressing cells were selected with 0.6 mg/ml geneticin and cloned by limiting dilution. Ld-expressing 721.221 lines have been described previously (12). The tapasin-deficient mouse ear fibroblast line (Tpn-/- MEF) was derived from tapasin knockout mice (6), and a clonal line was obtained by limiting dilution for use in these studies (called Tpn-/- 3.5). Transient transfections were performed using FuGene6 (Roche Diagnostics, Indianapolis, IN). All cells were maintained in RPMI 1640 (Life Technologies) supplemented with 10% bovine calf serum or 10% FCS (for the Tpn-/- MEF line; HyClone Laboratories, Logan, UT), 2 mM L-glutamine, 0.1 mM nonessential amino acids, 1.25 mM HEPES, 1 mM sodium pyruvate, and 100 U/ml penicillin/streptomycin (all from the Tissue Culture Support Center, Washington University School of Medicine). Where appropriate, geneticin was added to a final concentration of 0.6 mg/ml.
Peptides
The M3-binding peptides used in this study were: LemA (fMIGWII) and Fr38 (fFMIVIL) from Listeria monocytogenes (25, 29); ND1 (fMFFINIL) from murine NADH dehydrogenase (30); and TB6 (fMFFLDA; T. Chun and C.-R. Wang, manuscript in preparation). The Ld peptide p29 (YPN VNIHNF) was used (31). All peptides were synthesized using F-moc solid-phase chemistry (32) on an Applied Biosystems (Foster City, CA) model 432A peptide synthesizer. Purity was assessed by reverse-phase HPLC and mass spectrometry. For all experiments described in this study, peptides were dissolved in DMSO at a concentration of 10–20 mM to create stock solutions from which appropriate dilutions were obtained. The M3 peptides require DMSO solubilization, and the class Ia peptides were also dissolved in DMSO for consistency.
Antibodies
mAb 64-3-7 (27) is specific for open forms of
Ld and epitope-tagged molecules and was used for
immunoprecipitation, immunoblot, and flow cytometry. mAb 30-5-7
recognizes fully assembled, 2m-associated
Ld (33) and was used for
immunoprecipitations. mAb130 (22) is specific for M3 and
was used for flow cytometry and immunoprecipitation. Murine TAP
precipitations and blots were performed with rabbit antiserum 502
(34). The following Abs were used for immunoblot: rabbit
anti-human TAP (35); rabbit anti-human tapasin
(12); mAb BBM.1 (anti-human
2m) (36); rabbit anti-murine
tapasin (2668; M. R. Harris, L. Lybarger, Y. Y. L. Yu, N. B. Myers, and
T. H. Hansen, unpublished observations); rabbit anti-murine
2m (1419; M. R. Harris, L. Lybarger, Y. Y. L.
Yu, N. B. Myers, and T. H. Hansen, unpublished data); chicken
anticalreticulin serum (Affinity BioReagents, Golden, CO); and
rabbit anticalnexin (StressGen, Victoria, BC, Canada).
Flow cytometry
All flow cytometric analyses were performed using a FACSCalibur
(BD Biosciences, San Jose, CA). Dead cells and debris were excluded
from analysis on the basis of forward angle and side scatter light
gating. A minimum of 10,000 gated events was collected for analysis.
Data were analyzed using CellQuest software (BD Biosciences). For
surface staining, 
5 x 105 cells per
sample were incubated on ice in microtiter plates with culture
supernatant from the appropriate hybridoma. After washing,
fluorochrome-conjugated secondary Ab was added. To visualize mAb130
staining, FITC-labeled goat anti-Armenian hamster IgG was used
(Jackson ImmunoResearch, West Grove, PA) in most cases. For two-color
analyses involving YFP fluorescence, mAb130 staining was visualized
using PE-conjugated mouse anti-hamster IgG (BD PharMingen, San
Diego, CA). YFP fluorescence was collected in the FITC channel, and
electronic compensation, based on single-color controls, was applied to
segregate the two signals. For 64-3-7 staining, FITC-conjugated goat
anti-mouse IgG was used (ICN Pharmaceuticals, Costa Mesa, CA). For
intracellular staining of 64-3-7, cells were fixed and permeabilized in
PBS containing 1% paraformaldehyde, 1% BSA, and 0.5% saponin (all
from Sigma, St. Louis, MO) for 20 min on ice. Permeabilized cells were
stained in PBS containing 1% BSA and 0.5% saponin into which
FITC-conjugated 64-3-7 was diluted.
Immunoprecipitations and immunoblot
For coimmunoprecipitations, cells were lysed in TBS plus 1% digitonin (Wako, Richmond, VA) that contained a saturating concentration of precipitating Ab. After lysis for 30 min on ice, postnuclear lysates were incubated with protein A-Sepharose (Amersham Pharmacia Biotech, Uppsala, Sweden) for 1 h. Beads were washed four times in TBS plus 0.1% digitonin, and bound proteins were eluted by boiling in SDS sample buffer. Immunoblot was performed following SDS-PAGE separation of precipitated proteins and transfer to Immobilon P membranes (Millipore, Bedford, MA). Membranes were blocked (1 h to overnight) with PBS plus 10% dried milk plus 0.05% Tween 20. Primary Abs were added and incubated for 1 h, followed by washing in PBS plus 0.05% Tween 20. As a second step, membranes were incubated for 1 h with biotin-conjugated anti-mouse or anti-rabbit IgG (Caltag Laboratories, South San Francisco, CA). After washing, HRP-conjugated streptavidin (Zymed, San Francisco, CA) was added for 1 h, followed by three washes. Specific proteins were visualized by chemiluminescence using the ECL system (Amersham, Boston, MA).
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Studies of class Ia biosynthesis have been aided greatly by the availability of serological reagents that recognize H chain conformations that change during assembly (27, 37). For example, mAb 64-3-7 defines a unique conformation of Ld H chains that exists before peptide binding and after peptide dissociation, which we term the open conformation (26). Recently, we have demonstrated that this open conformation is defined primarily by the orientation of amino acid side chains at H chain residues 48 and 50 that undergo a secondary structure alteration coincident with high-affinity peptide binding. Class I H chain is in the open conformation when associated with the peptide-loading complex and, accordingly, 64-3-7 can be used to coprecipitate all known members of the loading complex. Our laboratory has developed a strategy by which the 64-3-7 epitope can be transferred with limited mutagenesis to other class I molecules and, thus, can be used to tag their respective open conformers (12, 26). We considered this a powerful tool to study the biosynthesis of M3 and how it compares with the biosynthesis of various class Ia molecules naturally expressing (e.g., Ld) or tagged with the same epitope. Using this approach, mAb 64-3-7 could be used to compare peptide-induced conformational changes in the ER and at the cell surface, of M3 vs class Ia molecules, and determine how these differences correlate with observed differences in their respective molecular chaperone interactions. In regard to studies of M3, this strategy was particularly efficacious owing to the comparisons that can be made with another mAb (130) described recently that recognizes M3 irrespective of its conformation (22).
Expression and detection of etM3
A point mutation was introduced into the M3 cDNA (R48Q) that
converted its sequence to the 64-3-7 epitope sequence found in
Ld. etM3 was transfected into a murine fibroblast
line (DAP-3), and stable etM3-expressing clones were identified by
intracellular staining with 64-3-7. Intracellular staining revealed a
sizable pool of etM3 (Fig. 1
A)
and demonstrated that the epitope transfer was successful. This pool of
64-3-7+ etM3 molecules was markedly decreased
when DAP-3.etM3 cells were cultured with an M3-binding peptide ligand,
Fr38. Thus, intracellular 64-3-7+ etM3 molecules
are highly peptide accessible. Either before or after peptide
treatment, no 64-3-7+ etM3 molecules were
detected at the cell surface. However, peptide treatment did result in
the expression of high levels of etM3 detected with mAb130. It should
be noted that mAb130 also detected a low level of endogenous, wild-type
M3 molecules expressed on the surface of DAP-3 cells following
induction (Fig. 1B, right panel). The fact that
mAb130 staining was 
1.5 logs higher on peptide-treated DAP-3.etM3 vs
DAP-3 demonstrated that the transfected etM3 gene is overexpressed
relative to the endogenous M3 gene in this cell line. These findings
indicate that open forms of etM3, as accurately defined by mAb 64-3-7,
can be detected intracellularly, but not at the cell surface either
before or after peptide treatment. In support of this conclusion,
pulse-chase experiments demonstrated that 64-3-7-reactive etM3 remained
endo H sensitive (hence, ER resident) after peptide treatment (data not
shown). By contrast, mAb130 reacts with M3 or etM3 surface molecules
after peptide treatment. The fact that surface etM3 molecules were only
detected with mAb130 and not mAb 64-3-7 provides compelling evidence
that they are indeed peptide occupied.
|
The detection by mAb130 of high levels of peptide-induced etM3 on
DAP-3.etM3 cells allowed us to compare the peptide-binding specificity
of etM3 molecules with that previously determined with wild-type M3
molecules. Two high-affinity M3 ligands (LemA and Fr38), one
intermediate binder (ND1), and one weak binder (TB6) were tested for
their ability to induce surface expression of etM3. As shown in Fig. 2, the relative binding affinities of the
different peptides for etM3 matched those that were reported for
wild-type M3 using the same assay (22) (T. Chun and C.-R.
Wang, manuscript in preparation). Therefore, transfer of the 64-3-7
epitope tag into M3 did not alter its peptide-binding specificity, a
conclusion supported by our extensive studies of peptide binding to
epitope-tagged forms of Kd,
Kb, and HLA-B27 (Refs. 12, 26 ,
and data not shown).
|
, acid stripping resulted in a severe
reduction in the detection of mAb130-reactive etM3 molecules and a
sharp increase in the detection of 64-3-7-reactive etM3. These findings
provide additional evidence that the 64-3-7 epitope tag can be used to
identify etM3 molecules that are truly peptide empty. What is less
clear is why such peptide-empty forms of etM3 were not also reactive
with mAb130, since mAb130 detects M3 molecules by immunoblot, and can
immunoprecipitate nascent M3 molecules derived from the ER that are
capable of binding exogenous peptide (22). Perhaps the
mAb130 epitope is especially acid sensitive, whereas the 64-3-7 epitope
is not. In any case, the obvious differences in the reactivity of mAb
64-3-7 vs 130 with etM3 molecules before and after acid stripping
demonstrated the utility on using both of these mAb to identify
distinct conformational intermediates during M3 biosynthesis.
|
Although expression of M3 is largely TAP dependent, it was not
known whether M3 associates physically with the loading complex. To
examine the potential interaction of M3 with members of the loading
complex, coimmunoprecipitations were performed. etM3 was precipitated
with 64-3-7, and the resulting precipitates were blotted to detect the
presence of ER chaperones. As a control, the same precipitations were
performed with Ld, which does associate with the
loading complex. The results demonstrated that M3 readily associates
with members of the loading complex, including TAP, tapasin,
calreticulin, and 2m (Fig. 4). The levels of each of these molecules
that coprecipitated were comparable with those obtained with
Ld. These chaperones were not detected in
precipitates of 30-5-7, an Ab that recognizes folded
Ld only, as expected. In addition, the overall
levels of open H chain were similar in Ld- and
etM3-expressing cells. Therefore, comparable levels of
Ld and etM3 H chains resulted in comparable
levels of steady-state association of each of these molecules with the
peptide-loading complex. It is noteworthy that these relative levels of
loading complex association appear to be typical of class I molecules
in transfected cells, as was shown by comparison of
Ld with etKb and
etKd (12). These results demonstrate
that M3 is capable of association with the peptide-loading complex, and
that the levels of association are similar between M3 and mouse class
Ia molecules. The fact that the restricted peptide pool of M3 does not
result in significantly higher steady-state association of M3 with the
loading complex, compared with class Ia molecules, suggests that
intrinsic properties of M3 may influence its ability to bind
TAP/tapasin complexes. This conclusion is strongly supported by studies
of M3 expression in human cell lines (see below).
|
M3 molecules are unique among TAP-associated molecules, since they
are only expressed at a high level in the presence of exogenous
peptides (21, 22) (Fig. 1
). Binding of peptide to M3
appears to occur almost exclusively in the ER, as suggested by the loss
of endo Hs forms of M3 H chains after peptide
induction (Refs. 22, 39 , and data not shown). In
contrast, class Ia molecules can undergo substantial peptide exchange
at the cell surface (40, 41), and peptide binding to both
endo Hs and endo Hr forms
of class Ia molecules such as Ld or
Kb has been observed when cells are incubated
with exogenous ligands (40, 41, 42). This unique aspect of M3
raised the possibility that we could assay for peptide-induced release
of M3 from TAP using live cells cultured with peptide. Initial
demonstrations of peptide-induced release of class Ia molecules from
TAP used permeabilized cells to better deliver peptides to the cytosol
(35, 43). However, with M3, peptide uptake by live cells
is clearly sufficient to promote a significant drop in the
intracellular pool of peptide-accessible M3 molecules (22, 39 , and Fig. 1). Thus, it was of interest to determine how
peptide treatment of nonpermeabilized cells affected the steady-state
levels of TAP-M3 association and how these findings compared with class
Ia molecules. To address these questions, anti-TAP immunoprecipitations
were performed with etM3- or Ld-expressing cells
cultured with or without exogenous peptide. These precipitates were
then blotted for H chain to determine the extent of peptide-induced TAP
release.
Fig. 5 demonstrates that the
TAP-associated pool (and by extension, the loading complex-associated
pool) of etM3 was markedly reduced by overnight incubation with
specific peptide (>10-fold). By comparison, exogenous peptide had only
a modest and transient TAP-release effect on Ld.
Multiple Ld-binding peptides have been tested at
high concentrations, and all failed to induce near total release of
Ld from TAP, including murine CMV, QL9,
and p29, reported to be high-affinity ligands (Refs. 31, 44, 45 , and data not shown). Indeed, the experiment shown in this
study is the highest level of TAP release of Ld
that we have observed (2-fold), and was obtained using a peptide,
p29 (31), which has the highest affinity of any
Ld ligands that we have tested (46).
In addition, our biochemical data on the extent of
Ld release from TAP correlate well with values
determined using fluorescence measurements of class I molecule
(Ld and Kb) diffusion in
membranes following peptide addition (47, 48). Thus, these
findings suggest that peptide-binding affinity alone does not explain
the incomplete dissociation of Ld from the
loading complex, compared with M3. The disparity between
Ld and M3 could be due to a unique
uptake/delivery pathway for M3 ligands.
|
M3 appears to be unique to rodents, as no ortholog has been
identified in humans, nor has a functional homologue been described.
Therefore, if M3 requires any specialized biosynthetic processes not
required for class Ia expression, then human cells may be incapable of
supporting M3 expression. To test this, etM3 was expressed in two human
cell lines, B lymphoblastoid cell line 721.221 (.221; 28) and
HeLa (cervical carcinoma). Transfectants of each line were selected
that contained sizable pools of intracellular open (64-3-7-positive)
etM3, and neither line expressed M3 at the surface (Fig. 6). However, incubation of these cells
with formylated peptide revealed a surprising difference between the
two transfectants. Although HeLa.etM3 cells exhibited robust surface
induction of M3 at levels similar to mouse L cells, .221.etM3 failed to
express etM3 at the surface subsequent to peptide addition. Thus, human
cells appear to possess all of the components necessary to support M3
expression, although the .221 line is defective in this process. It is
noteworthy that murine class Ia genes can be expressed at high levels
in .221 cells (12, 14), indicating an M3-specific defect
with this cell line.
|
A. Although the
64-3-7-positive form of the mouse class Ia molecule
(Ld) associated well with members of the human
loading complex, M3 was associated relatively weakly with each
chaperone, despite the similar levels of open etM3 present in all of
the lines. This was the case with either the HeLa cells (which can
express M3 at the surface) or .221 cells (which cannot express M3 at
the surface). For example, the ratio of open H chain signal:TAP signal
(determined by quantitation of the image shown in Fig. 7A)
was 0.67 for Ld in .221 cells, 3.8 for etM3 in
HeLa cells, and 2.6 for etM3 in .221 cells. Given the similar chaperone
association profiles for M3 in the HeLa and .221 cells, the lack of
induction in .221 cells cannot be explained by a failed association
with the human loading complex. The finding that open forms of etM3
showed similar 2m association levels to open
Ld in .221 cells indicates that etM3 is not
grossly misfolded in the .221 line.
|
B). Under these assay conditions, Fr38 peptide
was able to induce substantial folding (loss of 64-3-7 reactivity) of
etM3 in a concentration-dependent manner in lysates from HeLa and
DAP-3. The levels of etM3 precipitated with mAb130 remained constant,
indicating that this mAb130 reacts with M3 irrespective of its folding
status, and provides clear evidence that peptide converts etM3 to a
64-3-7-negative form. By contrast, peptide had no effect on etM3 from
the .221 lysate at the highest concentration tested. This finding
predicts a defect in peptide acquisition by M3 in these cells that lies
downstream of peptide entry/uptake from the culture medium. Although
the nature of this defect is at present unknown, it is intriguing to
speculate that it might involve a heat shock protein (HSP). HSPs have
been recently implicated in peptide stabilization and delivery to MHC
class I molecules (49). Perhaps a chemical property unique
to M3 ligands, such as their shorter length or high degree of
insolubility, might render them more dependent upon a particular HSP
than class Ia peptide ligands. This may also help explain the dramatic
TAP release of M3 caused by exogenous peptides, if a HSP capable of
interacting with the loading complex was involved in the delivery of
peptides to M3. Regardless of the exact nature of the defect in .221
cells, these observations revealed that a factor(s) dispensable for
class Ia expression is required for M3 folding. M3 expression in tapasin-deficient cells
The finding that etM3 was poorly associated with the loading complex in HeLa cells, yet was retained within the cell and could be surface induced with exogenous peptide, raised the possibility that association with the loading complex was not required for these processes. To test this hypothesis more rigorously, we examined M3 expression in a fibroblast cell line derived from tapasin-deficient mice (6). M3 represented a uniquely powerful system to test the ER retention function that has been ascribed to tapasin (10, 13, 14), since the available pool of endogenous ligands is so limited, resulting in an extensive intracellular pool of open M3 molecules. In the tapasin-deficient background, we could test whether tapasin was required for folding of M3 leading to surface induction, and whether open forms were efficiently retained within the cell.
To perform these analyses, we constructed an etM3/YFP chimera,
with YFP linked to the cytoplasmic tail of etM3. Such fluorescent
protein fusions have been made with class Ia molecules without adverse
effects on their processing, expression, and association with the
loading complex (47, 48). Indeed, the etM3/YFP chimeric
molecule associates with the peptide-loading complex (data not shown),
and its expression is strictly regulated by peptide availability,
identical to native M3 (Fig. 8 and data
not shown). The YFP tag permits the unequivocal identification of
etM3-transfected cells by flow cytometry, and the surface expression of
etM3 can be monitored simultaneously using mAb130 staining. The
tapasin-deficient cell line (Tpn-/- 3.5) was
transfected transiently with the etM3/YFP construct alone or
cotransfected with a mouse tapasin (Tpn) expression vector, and the
transfectants were subsequently incubated overnight with Fr38 peptide.
After peptide incubation, cells were analyzed for YFP expression and
mAb130 reactivity. Fig. 8 depicts the results from one representative
experiment (of four), and Table I
summarizes the data from all experiments. Since a fraction of the
transfectants expressed the etM3/YFP chimera, the surface expression
pattern of both the endogenous M3 and the transfected etM3 could be
evaluated, by gating analysis on the YFP- or
YFP+ fractions, respectively. Furthermore, the
intensity of the YFP fluorescence signal could be used to normalize the
samples for transfection efficiency.
|
|
). This was true even for
the cells that expressed very high levels of transfected etM3/YFP
(>1000-fold range of expression). These findings indicated that
tapasin was not required for intracellular retention of M3, and are
supported by studies of tapasin-deficient mice, in which M3 molecules
remained endo H sensitive (hence, ER resident) in the absence of
exogenous peptide.4 Another
striking finding of this experiment was that both endogenous and
transfected M3 could be induced at the cell surface by peptide in the
Tpn-/- cells. The level of induction of
endogenous M3 was comparable (5-fold) in these cells with that
observed with wild-type fibroblasts under the same conditions (data not
shown), and the transfected etM3 (YFP+ fraction)
was induced to still higher levels. This demonstrated that tapasin was
not required for intracellular retention or induced surface expression
of M3 in this cell line. However, tapasin did enhance peptide-induced
expression of M3, since cotransfection of tapasin with etM3 increased
the surface induction of etM3 2- to -3 fold (Fig. 8, middle
panels, and Table I). Given the ER localization of tapasin, this
provides further evidence that exogenous peptide binding by etM3
molecules occurs primarily within the cell, a finding strongly
supported by data presented earlier (Figs. 1 and 5) and previous
reports (22, 39). Tapasin has been shown to stabilize the TAP heterodimer, leading to increased peptide transport (10), and fibroblasts from Tpn-/- mice have a drastic TAP transport defect (6), presumably due to lower levels of TAP expression. To distinguish between the TAP-stabilizing effects of tapasin and a more direct role of tapasin in M3 expression, Tpn-/- cells were transiently transfected with a soluble murine tapasin construct. Soluble tapasin (lacking the transmembrane and cytosolic portions) binds to class I H chain, but cannot bridge the H chain to TAP or stabilize the TAP heterodimer (10). Importantly, soluble tapasin also fails to retain class I molecules (slow the intracellular maturation kinetics) (10, 14). Analysis of cells that were cotransfected with the soluble tapasin and etM3/YFP constructs revealed that soluble tapasin was able to enhance M3 surface expression to levels similar to those observed with the full-length tapasin construct. Immunoblot analysis of the transfectants revealed that both the full-length and soluble tapasin constructs were expressed at comparable levels (data not shown). These data support a direct role for tapasin in peptide-induced H chain folding independent of intracellular retention.
The efficient retention of M3 in the absence of tapasin contrasts with surface expression of Kb in insect cells. Using a serological approach similar to that used in this study, Schoenhals et al. (13) showed that in the absence of tapasin, high levels of open Kb conformers were expressed at the cell surface. However, when tapasin and Kb were coexpressed, open Kb conformers were retained intracellularly. In addition, studies using mammalian cells have demonstrated that full-length tapasin decreases the maturation kinetics of class Ia molecules in tapasin-deficient .220 cells (10, 14). Thus, the lack of a requirement for tapasin in the ER retention of M3 appears to conflict with reported findings for class Ia molecules. Furthermore, using calnexin-deficient cells (51, 52), we have found that calnexin is also not required for efficient intracellular retention of M3 (data not shown). Collectively, our data provide compelling evidence that tapasin directly facilitates peptide binding to M3 by a process independent of the role of tapasin in ER retention and TAP stabilization. How might our findings with M3 relate to ER retention of class Ia molecules? It should be noted that evidence for the role of tapasin in ER retention is based on studies using invertebrate cells (13) or human .220 cells (which are heavily mutagenized) (10, 14). Thus, either of these cell types may lack additional quality control molecules that are normally present in mammalian cells. Furthermore, in tapasin-deficient mouse cells, class I molecules escape the ER in a peptide-accessible form, suggesting that they may contain suboptimal peptides (6). The apparent difference between M3 and class Ia molecules, in terms of a requirement for tapasin in the ER retention of open conformers, might be attributable to their respective peptide pools. More specifically, tapasin may not be required for ER retention of peptide-empty forms of either class Ia or M3 molecules. Rather, class Ia molecules that leave the ER in the absence of tapasin could be fully conformed with peptide, but these peptides are of generally lower quality/affinity, whereas endogenous ligands of sufficient affinity to induce folding and ER egress of M3 are lacking. In any case, our findings with soluble tapasin point to a direct role of tapasin in peptide binding to M3, and our findings with human .221 cells indicate that M3 expression requires a unique factor that is dispensable for the expression of class Ia molecules.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ted H. Hansen, Department of Genetics, Box 8232, Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110. E-mail address: hansen{at}genetics.wustl.edu
3 Abbreviations used in this paper: 2m, 2-microglobulin; ER, endoplasmic reticulum; LCL, lymphoblastoid cell line; et, epitope tagged; HSP, heat shock protein; YFP, yellow fluorescent protein.
4 T. Chun, A. G. Grandea III, L. Lybarger, J. Forman, L. Van Kaer, and C.-R. Wang. Functional roles of TAP and tapasin in the assembly of M3-N-formylated peptide complexes. Submitted for publication.
Received for publication April 12, 2001. Accepted for publication June 14, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
2-microglobulin complexes associate with TAP transporters before peptide binding. Nature 368:864.[Medline]
2-microglobulin antibody and its use in the genetic and biochemical analysis of major hisotcompatibility antigens. Eur. J. Immunol. 9:536.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. Tuli, M. Sharma, M. M. McIlhaney, J. E. Talmadge, N. Naslavsky, S. Caplan, and J. C. Solheim Amyloid Precursor-Like Protein 2 Increases the Endocytosis, Instability, and Turnover of the H2-Kd MHC Class I Molecule J. Immunol., August 1, 2008; 181(3): 1978 - 1987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Lou, G. Basha, R. P. Seipp, B. Cai, S. S. Chen, A. R. Moise, A. P. Jeffries, R. S. Gopaul, T. Z. Vitalis, and W. A. Jefferies Combining the Antigen Processing Components TAP and Tapasin Elicits Enhanced Tumor-Free Survival Clin. Cancer Res., March 1, 2008; 14(5): 1494 - 1501. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. W. Everett and M. Edidin Tapasin Increases Efficiency of MHC I Assembly in the Endoplasmic Reticulum but Does Not Affect MHC I Stability at the Cell Surface J. Immunol., December 1, 2007; 179(11): 7646 - 7652. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Rizvi and M. Raghavan Direct peptide-regulatable interactions between MHC class I molecules and tapasin PNAS, November 28, 2006; 103(48): 18220 - 18225. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Thammavongsa, G. Raghuraman, T. M. Filzen, K. L. Collins, and M. Raghavan HLA-B44 Polymorphisms at Position 116 of the Heavy Chain Influence TAP Complex Binding via an Effect on Peptide Occupancy. J. Immunol., September 1, 2006; 177(5): 3150 - 3161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Huang, S. Gilfillan, M. Cella, M. J. Miley, O. Lantz, L. Lybarger, D. H. Fremont, and T. H. Hansen Evidence for MR1 Antigen Presentation to Mucosal-associated Invariant T Cells J. Biol. Chem., June 3, 2005; 280(22): 21183 - 21193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Wang, R. Connors, M. R. Harris, T. H. Hansen, and L. Lybarger Requirements for the Selective Degradation of Endoplasmic Reticulum-Resident Major Histocompatibility Complex Class I Proteins by the Viral Immune Evasion Molecule mK3 J. Virol., April 1, 2005; 79(7): 4099 - 4108. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Li, B. A. Sullivan, C. J. Aldrich, M. J. Soloski, J. Forman, A. G. Grandea III, P. E. Jensen, and L. Van Kaer Differential Requirement for Tapasin in the Presentation of Leader- and Insulin-Derived Peptide Antigens to Qa-1b-Restricted CTLs J. Immunol., September 15, 2004; 173(6): 3707 - 3715. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Howarth, A. Williams, A. B. Tolstrup, and T. Elliott Tapasin enhances MHC class I peptide presentation according to peptide half-life PNAS, August 10, 2004; 101(32): 11737 - 11742. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. PAULSSON and P. WANG Quality control of MHC class I maturation FASEB J, January 1, 2004; 18(1): 31 - 38. [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]
|
|
|
|
|
|
M. J. Miley, S. M. Truscott, Y. Y. L. Yu, S. Gilfillan, D. H. Fremont, T. H. Hansen, and L. Lybarger Biochemical Features of the MHC-Related Protein 1 Consistent with an Immunological Function J. Immunol., June 15, 2003; 170(12): 6090 - 6098. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Park and K. Ahn An Essential Function of Tapasin in Quality Control of HLA-G Molecules J. Biol. Chem., April 11, 2003; 278(16): 14337 - 14345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.-J. Kang and P. Cresswell Calnexin, Calreticulin, and ERp57 Cooperate in Disulfide Bond Formation in Human CD1d Heavy Chain J. Biol. Chem., November 15, 2002; 277(47): 44838 - 44844. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Y. L. Yu, M. R. Harris, L. Lybarger, L. A. Kimpler, N. B. Myers, H. W. Virgin IV, and T. H. Hansen Physical Association of the K3 Protein of Gamma-2 Herpesvirus 68 with Major Histocompatibility Complex Class I Molecules with Impaired Peptide and {beta}2-Microglobulin Assembly J. Virol., February 22, 2002; 76(6): 2796 - 2803. [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]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
