| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
*
Division of Immunobiology, University of Bonn, Bonn, Germany;
Human Immunogenetics Laboratory, Imperial Cancer Research Fund, London, United Kingdom; and
Department of Cellular Biochemistry, The Netherlands Cancer Institute, Amsterdam, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß heterodimers were
readily detected, no DR complexes with an (ß)2-chain
composition could be identified. Two mAbs (L243 and D1-12)
immunoprecipitated high m.w. DR complexes suspected to be superdimers.
However, biochemical analysis revealed that, rather than superdimers,
these were SDS-stable complexes of DR in combination with the Abs.
Thus, previous observations of HLA-DR superdimer bands may also reflect
complexes of DR molecules with bound Ab. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß heterodimers formed a dimer of dimers, or superdimer
(1). This result was unexpected because, in numerous previous
biochemical studies, a tetrameric class II complex had not been
detected. There has been great interest in potential functions of
superdimers in T cell Ag recognition, and a variety of cellular roles
have been proposed for them (2, 3). T cell Ag recognition takes place
by multiple interactions between the TCR and peptide-loaded MHC as well
as accessory molecules that stabilize the binding between APC and T
cells. It is tempting to speculate that two associated class II dimers
loaded with matching antigenic peptides may enhance T cell responses by
cross-linking of two TCR with identical specificity. In this way, a
superdimeric structure could facilitate specific recognition of Ag
despite the weak interactions of some of the subunits involved. In
addition, signal transduction could be amplified in T cells induced by
dimeric class II heterodimers. However, the possibility that two MHC
molecules in a dimer present both the same peptide may be achieved only
at high Ag concentrations (4). In spite of the theoretical
attractiveness of class II superdimers, it is possible that the
tetrameric HLA-DR1 complex is a crystallization-packing phenomenon that
does not reflect physiological conditions. Additional evidence for
superdimers comes from the use of particular Abs in
immunoprecipitation. Mouse class II H2E
(IE)4 Ags were immunoisolated with the mAb
Y17 (5, 6). A high m.w. band that resists SDS treatment was suggested
to represent a superdimeric structure of mouse class II molecules.
Similar studies with HLA-DR molecules have also been reported with
D1-12 mAb (7). We set out to determine whether superdimeric structures
exist on human APCs. We have also studied the nature of high m.w. DR
complexes in the presence of L243 and D1-12 mAbs. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Proteins were separated on 10 or 13% polyacrylamide SDS gels. Two-dimensional electrophoresis (nonboiled/boiled) was conducted as previously described (8). Samples treated at room temperature with reducing or nonreducing SDS sample buffer were separated in the first dimension in rod gels. After electrophoresis, these gels were boiled for 5 min in sample buffer and fixed on the top of an SDS gel, followed by separation in the second dimension.
mAbs and immunoprecipitation
The DR-specific mouse mAbs L243, D1-12, 1B5, I-80AG11, I-247, I-251SB, and ISCR3 have been described (9, 10, 11, 12, 13, 14). The mAb PA1 against the human transferrin receptor was a kind gift from Dr. G. Moldenhauer (Heidelberg, Germany). The goat anti-mouse Fab Ab (code No. 115-005-072) was purchased from Dianova (Hamburg, Germany). This Ab is affinity purified and reacts with mouse Ig light chain but not with the Fc portion. For immunoprecipitation, pellets of 2 x 106 cells were suspended in 450 µl TBS (pH 7.4) and lysed by addition of 50 µl 10% Nonidet P-40. DNA and debris were removed by high speed centrifugation, and the lysate was precleared for 2 h at 4°C with Sepharose CL4B. The lysate was then mixed with 50 µl 20x concentrated hybridoma supernatant containing mAbs and 10 µl protein A-Sepharose and gently mixed at 4°C overnight. The protein A-Sepharose beads were washed three times with 500 µl of 0.25% Nonidet P-40 in TBS and subsequently incubated for 2 h with SDS sample buffer followed by electrophoresis. Protease inhibitors PMSF (1 mM) and Trasylol (1:1000) were included in all experiments.
Metabolic labeling and cell lines
The human B lymphoma cell lines Raji and JOK-1 were maintained in RPMI supplemented with 10% FCS. Cells were washed intensively with medium lacking methionine and pulse labeled for 15 min with 50 µCi [35S]methionine (1 Ci = 37 GBq) (Amersham, Braunschweig, Germany). After addition of medium complemented with unlabeled methionine, cells were maintained for an overnight chase period. They were then washed, and the cell pellet was either stored at -20°C or lysed for immunoprecipitation.
Preparation of microsomes and coupling of Abs to Sepharose
Two times 108 cells were washed and incubated in a hypotonic buffer (10 mM Tris (pH 7.8), 30 mM MgCl2) on ice and subsequently sheared by douncing to separate membranes from nuclei. Nuclei were removed by centrifugation for 15 min at 1500 x g. To remove the cytosolic phase, the supernatant was centrifuged for 1 h at 40,000 x g, and microsomes were obtained as a small white pellet. Protease inhibitors Trasylol (1:1000) and PMSF (1 mM) were present in all procedures. Abs purified with protein A-Sepharose from hybridoma supernatant were precipitated with (NH4)2S04 and dialyzed against coupling buffer (0.1 M NaHCO3, pH 8.3). Cyan-activated Sepharose 4B beads (Sigma, St. Louis, MO) were treated with 1 mM HCl and incubated overnight at 4°C with the mAb solution. Then beads were incubated with blocking buffer (0.2 M glycin) for 2 h at room temperature followed by two washing steps with 0.1 M acetate buffer (pH 4) containing 0.5 M NaCl and with coupling buffer.
Western blotting
Cell lysates or microsome preparations equivalent to 2 x
106 cells were incubated with SDS sample buffer for 2
h at room temperature and subsequently separated by PAGE as described
above. The gels were calibrated using prestained m.w. markers
(Amersham). Proteins were transferred to polyvinylidene
difluoride membranes (Immobilon, Millipore, Bedford,
MA), and the filter was blocked with 5% skimmed milk powder
dissolved in 0.1% Tween 20/PBS for 1 h and then incubated for
1 h with a 1:500 dilution of 1B5 (anti DRAb) hybridoma
supernatant. After three washes with 0.1% Tween 20/PBS, the filter was
incubated for 1 h with 1:2000 dilution of horseradish
peroxidase-conjugated rabbit anti-mouse Ig (Dako, Hamburg,
Germany, 1 mg/ml), followed by detection of bound Ab by enhanced
chemiluminescence (Amersham).
Preparation of F(ab')2 fragments
L243 and D1-12 mAbs were isolated by affinity chromatography with protein A-Sepharose from hybridoma supernatant and subsequently dialyzed against 100 mM sodium acetate buffer (pH 5.5). Cysteine and EDTA were added to final concentrations 50 mM and 1 mM, respectively. Ten micrograms pepsin per milligram of Ab was added, and the solution was incubated at 37°C. Optimal digestion was obtained after 2 h (L243) or 8 h (D1-12) of incubation. To stop the digestion, iodacetamide was added to a final concentration of 75 mM, and incubation at room temperature was proceeded for another 30 min. The digestion then was gently rotated overnight with protein A-Sepharose to remove Fc fragments and undigested Ig. F(ab')2 fragments were analyzed by SDS-PAGE, and their size was determined after Coomassie blue staining.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We inspected the B lymphoma cell line Raji for the presence of
heat-labile complexes of DR polypeptides. Fig. 1
A shows a Western
blot of microsomes prepared from Raji cells and, for comparison, a DR
immunoprecipitate from a lysate of the same cells. Samples were treated
at room temperature with reducing SDS sample buffer (lanes
1 and 2) or boiled for 5 min (lanes 3
and 4) and separated by SDS-PAGE. After transfer of the
proteins, the filter was probed with the DR-chain-specific mAb 1B5.
The microsomal preparation from Raji cells exhibited two bands
corresponding to free DR-chain and to a heat-labile ß
heterodimer at about 65 kDa (lanes 1 and
3). These two bands also appeared in DR immunoprecipitates
(lanes 2 and 4). However, no band above 70
kDa, which would correspond to a superdimer, was detected in the
membrane preparation from Raji cells (lane 1). To
achieve denaturation of a potential high m.w. DR band, the same filter
was subsequently boiled for 3 min and again probed with 1B5. No
additional band was detected (data not shown). Probing of a filter
containing the same proteins with rabbit anti-mouse Ig conjugated
with peroxidase was used to indicate the position of Ig chains
(lanes 5–8).
|
subunit
with mAb 1B5. Lysates of the B lymphoma cell line JOK-1 were treated at
room temperature with reducing sample buffer and subjected to SDS-PAGE
in rod gels. After electrophoresis, these gels were incubated for 5 min
at 95°C to dissociate heat-labile complexes. The rod gels were fixed
on top of an SDS gel and separated, in the second dimension by PAGE.
Subsequently, the proteins were transferred to polyvinylidene
difluoride membranes and probed with 1B5. In this two-dimensional
separation, subunits of high m.w. DR complexes are located below a
diagonal (Fig. 1
B). Prestained m.w. markers, which were
separated in the same gel, calibrate the first and second dimensions
and indicate that DR-chains, corresponding to the position of ß
heterodimers as well as to free -chains, were found. However, no
DR subunit in the position of a four-chain DR complex was detected.
This result indicates the absence of heat-labile superdimer complexes
of DR molecules in untreated lysates of JOK-1 cells. Similar results
were obtained in other B lymphoma cell lines and in total lysates, in
addition to microsomes. Abs immunoisolate SDS-resistant high m.w. DR complexes
Recently, the isolation of high m.w. DR complexes suspected to be
DR superdimers was described using mAb D1-12 (7). As shown in Fig. 2, two mAbs, D1-12 and L243,
immunoprecipitated a band that ran with a low mobility and remained
near the top of the gel (lanes 1 and 5).
The high m.w. band was not immunoisolated from 15-min pulse-labeled
cells, suggesting that maturation of DR molecules is necessary for its
formation (data not shown). Incubation at 50°C yielded SDS-stable
ß dimers (apparent m.w. 65 kDa) but no high m.w. bands, indicating
that the DR dimer is more stable than the complexes (not shown). After
boiling, the high m.w. DR band dissociated to free - and ß-chains
(Fig. 2, lanes 2 and 6). To test whether
the high m.w. band represents a superdimeric DR complex, class II
molecules were isolated with D1-12 and L243 mAbs coupled to Sepharose
(lanes 3, 4, 7 and
8). Ab covalently bound to Sepharose cannot be removed by
SDS treatment. When DR molecules were immunoisolated using Abs coupled
in this way and the probes were treated at room temperature with SDS
sample buffer, almost no DR ß-chains and no high m.w. DR complexes
were observed on the gel (lanes 3 and
7). The reactivity of the L243 and D1-12 Sepharose with DR
is shown after boiling the probes in SDS sample buffer
(lanes 4 and 8). This treatment
dissociates DR - and ß-chains from the covalently bound Abs. These
experiments show that the appearance of high m.w. DR complexes is
dependent upon the soluble Ab substrate. We compared DR
immunoprecipitates of three mAbs, ISCR3, I-251SB and I-80AG11, to
immunoisolates with D1-12 (Fig. 3). While
D1-12 yields a high m.w. band, the other mAbs isolate a 55-kDa band
that corresponds to SDS-resistant ß dimers bound to peptide (14).
Two additional DR mAbs, 2.06 and I-247, also exclusively
immunoprecipitate the 55-kDa band (not shown). This suggests that the
appearance of the high m.w. band obtained with D1-12 and L243 mAbs is
an unusual feature of these two mAbs. We wanted to determine the
approximate size of the high m.w. DR bands. DR immunoprecipitates were
separated on a large gel with a gradient of 7.5–12% of acrylamide
(Fig. 4). In this gel, the separation of
the high m.w. bands is enhanced. Lane 1 shows heat-labile DR
dimers that were immunoisolated by mAb ISCR3. The high m.w. bands
isolated by D1-12 and L243 (lanes 3 and
5) run below the transferrin receptor dimer (180 kDa) that
was separated in lane 7. The D1-12- and L243-produced bands
show an unequal size, with estimated m.w. of approximately 145 kDa and
160 kDa. This variant size of D1-12 and L243 bands suggests that the
Abs cause the different mobilities of the high m.w. bands.
|
|
|
We considered the possibility that both Abs L243 and D1-12
strongly bind to DR molecules and are only partially dissociated from
the Ag by SDS treatment at room temperature. The approach was to
prepare F(ab')2 fragments of the Abs and to look for a
difference in size of the high m.w. DR complexes compared with when
complete Ab was used. DR molecules from 15-min pulse
[35S]methionine-labeled Raji cells, subsequently chased
overnight, were immunoprecipitated with F(ab')2 fragments
from L243 and D1-12. Since the fragmented Abs do not bind to protein
A-Sepharose, a sandwich Ab with specificity for Fab was used to
immunoisolate DR immunocomplexes. F(ab')2 fragments
resulted in SDS-resistant DR dimers but not high m.w. bands
characteristic of superdimers and some aggregates that appear at the
top of the gel (Fig. 5A,
lanes 3 and 7). The aggregates presumably
are caused by cross-linking of DR immunocomplexes by the bivalent
anti-Fab Abs. After boiling of the samples, the amount of ß
monomers increased (lanes 4 and 8).
This result was unexpected, because L243 and D1-12
F(ab')2-DR complexes should be stable in SDS. An
explanation of this result is given in Fig. 5B. By using the
anti-Fab sandwich Ab for isolation of DR immunocomplexes with
complete L243 and of D1-12 mAbs, again only DR dimers and some
aggregates but no high m.w. bands were obtained (lanes
2 and 6). This result suggests that the
polyclonal sandwich Abs either cross-link DR complexes or abrogate the
SDS stability of the Ab-DR complex that yields DR dimers.
|
). The DR-specific mAb I-247
immunoprecipitated an SDS-resistant DR ß dimer (lane
3) that dissociated after boiling to free - and ß-chains
(lane 6). This mAb was employed to isolate
F(ab')2-DR complexes from D1-12 and from L243 fragments
bound to DR molecules (lanes 2 and
8). The high m.w. F(ab')2-DR complex isolated in
lanes 2 and 8 was shifted by about 50 kDa
when compared with immunoprecipitates with complete mAbs D1-12 and L243
(lanes 1 and 7). The
F(ab')2-DR complex in lanes 2 and
8 dissociated after boiling (lanes 5
and 10). This experiment indicates that the mAbs D1-12 and
L243 or their F(ab')2 fragments are contained in the high
m.w. DR complexes and that this complex resists SDS treatment at room
temperature. The superdimer bands that recently have been described
thus may consist of DR-Ab complexes (7).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and ß-chains were
isolated. Thus, mAbs D1-12 and L243 recognize a maturation epitope of
DR molecules that is also present on the cell surface. We have presented two pieces of data that throw doubt on the existence of DR superdimer bands in acrylamide gels. 1) High m.w. bands of superdimer size were not found in the absence of Abs, at least in SDS 2) Two Abs that gave rise to high m.w. bands were found to bind to DR so strongly that they were dissociated only at high temperatures.
These results verify the observation made by Roucard et al. (7) that D1-12 mAb immunoprecipitates a high m.w. band. In contrast to these authors, we did not detect a superdimer band in SDS-PAGE-separated cell extracts by Western blotting. We give evidence that the high m.w. band immunoprecipitated with D1-12 and L243 Abs consists of DR heterodimers coupled to Ab. The recently reported isolation of superdimers from IE molecules with a mAb may reflect a similar phenomenon (5, 6). It is still unexplained why IE superdimer bands should be detected only by mAb Y17 and not by the great number of other IEk-specific mAbs. In addition, superdimer bands in other IE haplotypes should be identified if dimers of dimers are a substantial feature of class II molecules. Our experiments have concerned only human DR, and we cannot rule out that species differences exist.
Our findings suggest that some superdimers observed in crystals may
simply reflect constraints on packing in vitro. Recently reported
crystallization of IEk molecules supports this assumption
because there the pairs of class II dimers were in a completely
different orientation to those previously published by Brown et al.
(15). There remains a possibility that superdimers do have a
physiological role, but they are transiently formed in vivo only upon
interaction with TCR. Mutations of the two residues Glu88
and Lys111 of the DR-chain, which in the crystals form
two reciprocal salt bridges between the two dimers, were functionally
tested (16). Only the mutation of Glu88 abrogates Ag
presentation whereas mutation of Lys111 has no effect. In
conclusion, these data suggest that previous interpretation of DR
superdimers on gels may have in fact been due to DR-Ig complexes that
failed to be dissociated. Our data do not rule out the existence of
superdimers at the cell surface, such as recently have been detected
with single-particle fluorescence (17). There, it has been shown that
DR heterodimers and dimers of dimers may exist in a
temperature-dependent equilibrium. Recent experiments indicate that
recombinant dimeric class II molecules loaded with oligomeric ligands
cross-link TCRs and elicit an enhanced stimulation of T cells (18). In
addition, oligomeric class II/ligand complexes elevate activation of T
cells, suggesting that larger clusters of TCR might be functionally
active (19). Thus, class II superdimers or even larger class II
structures could facilitate the activation of T cells.
|
Footnotes |
|---|
2 Current address: Division of Immunology, Department of Pathology, University of Cambridge, Cambridge, United Kingdom
3 Address correspondence and reprint requests to Dr. Norbert Koch, Abteilung Immunbiologie, Universität Bonn, Römerstrasse 164, D53117 Bonn, Germany. E-mail address:
4 Abbreviations used in this paper: IE, mouse class II H2E; IM, immunoprecipitate.
Received for publication September 8, 1998. Accepted for publication January 14, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß)2 superdimers. J. Immunol. 161:2307. and ß subunits are present in the human Ia molecular pool. Proc. Natl. Acad. Sci. USA 78:4549.-chain subunits of human Ia alloantigens. Immunology 50:613.[Medline]
residues involved in dimers of HLA-DR molecules. J. Immunol. 155:1210.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  H. Li, Y. Zhao, Y. Guo, Z. Li, L. Eisele, and W. Mourad Zinc Induces Dimerization of the Class II Major Histocompatibility Complex Molecule That Leads to Cooperative Binding to a Superantigen J. Biol. Chem., March 2, 2007; 282(9): 5991 - 6000. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Muntasell, M. Carrascal, I. Alvarez, L. Serradell, P. van Veelen, F. A. W. Verreck, F. Koning, J. Abian, and D. Jaraquemada Dissection of the HLA-DR4 Peptide Repertoire in Endocrine Epithelial Cells: Strong Influence of Invariant Chain and HLA-DM Expression on the Nature of Ligands J. Immunol., July 15, 2004; 173(2): 1085 - 1093. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Lindstedt, N. Monk, G. Lombardi, and R. Lechler Amino Acid Substitutions in the Putative MHC Class II ""Dimer of Dimers"" Interface Inhibit CD4+ T Cell Activation J. Immunol., January 15, 2001; 166(2): 800 - 808. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
