| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Program in Immunology, Department of Pathology, Tufts University School of Medicine, Boston, MA 02111; and
Department of Pathology, Harvard Medical School, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
We show here that MMTV superantigen presentation does not require DM,
because transfection of the Mtv-7 sag gene into a
DR3-expressing cell line that is mutant for DMB (9) does
not affect functional recognition. To date, the best-characterized
protein that associates in unprocessed form with the class II 
heterodimer and also does not require DM for the association is Ii
(2, 3). The interaction of Ii with class II molecules
involves the peptide binding groove and takes place in the ER via a
portion of the lumenal domain of Ii, referred to as CLIP (class
II-associated Ii peptide) (10, 11). The MMTV superantigen
and Ii have structural and functional similarities: they both are type
II transmembrane proteins of similar size, and both are known to
associate with MHC class II inside the cell. In addition, we have now
identified in the protein sequence of the Mtv-7 superantigen
a strong class II peptide binding motif (CIIPBM) at a location
corresponding to CLIP in Ii. We implicate CIIPBM in the association of
MMTV superantigen with MHC class II molecules in an in vitro
translation system and in a functional superantigen presentation assay.
Our results suggest that these endogenously synthesized viral
superantigens can bind to MHC class II molecules during biosynthesis in
the ER in a manner similar to that of Ii.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
HLA-DR1 (12) and HLA-DR1 (13)
cDNAs were cloned into pSP72 (Promega, Madison, WI) behind the T7 RNA
polymerase promoter (14). In the Ii/Mtv-7 sag
constructs, IM1 and IM2, the amino-terminal portion of the Mtv-7
sag was replaced with the corresponding region of the Ii cDNA. IM1
contains an Ii cDNA replacement, corresponding to aa 1–106, which
includes the CLIP motif (MRMATPLLM), while construct IM2 contains an Ii
cDNA replacement, corresponding to aa 1–90, excluding the CLIP region.
Both constructs contain the Mtv-7 sag extracellular portion
from the transmembrane domain, including the CIIPBM (YNLNNSENS).
Construct m86IM2 is derived from IM2 by altering the position 1 (P1)
and P2 positions of CIIPBM by site-directed mutagenesis, Y
A and
NG, respectively, using Transformer (Clontech, Palo Alto, CA).
Nucleotides 256–261 of Mtv-7 sag (TACAAT) were altered to
GCCGGC, introducing an NaeI restriction site in m86IM2. This
same mutation was made in the wild-type Mtv-7 sag gene,
referred to as P1S7, for the transfection experiments (see
below), using the QuikChange site-directed mutagenesis kit (Stratagene,
La Jolla, CA). The P1S7 construct was subcloned into phAPr-1-neo
(15) to make phA-P1S7. For the in vitro translation
studies, the cDNAs of the Mtv-7 sag constructs were cloned
into a mouse encephalomyocarditis virus promoter-driven expression
vector (16).
Peptide scoring system
The peptide scoring system reflects the relative capacity of
nonameric peptides to bind to MHC class II molecules (17).
It is based on the hypothesis that this interaction is a result of all
side chain effects in any given peptide, and that the magnitude of the
side chain effect of a particular amino acid depends on its relative
position with regard to the anchor at P1. Because the P1 anchor
requires an amino acid with either a long aliphatic or an aromatic side
chain to interact with the deep pocket in the class II peptide binding
groove, only seven amino acids (F, I, L, M, V, W, and Y) can be
considered for this position. Thus, the protein (Mtv-7 sag, Ii, or
influenza virus hemagglutinin 306–318 peptide) (11) was
scanned for all potential P1 anchors, and then the adjacent 8 aa
residues (P2–P9) were located, and values obtained from side chain
scanning on DR1*0401 as defined by the algorithm (17)
were assigned to each amino acid. residue of the selected nonamer. The
sum of these values was the peptide score. A score of 2 or more
predicted class II binding.
In vitro translation and immunoprecipitation
The cDNAs were transcribed in vitro, either together or separately, using T7 RNA polymerase (Promega). The optimal amount of RNA for translation was determined empirically for each separate batch of RNA. In vitro translations were performed with rabbit reticulocyte lysate, not supplemented with DTT (Flexi; Promega), in the presence of canine pancreatic microsomal membranes (Promega). Translations were performed for 60 min at 30°C. After translation, microsomes were pelleted by centrifugation at 14,000 rpm for 10 min at 4°C and subsequently lysed in 200 µl of Nonidet P-40 lysis buffer (0.5% Nonidet P-40, 50 mM Tris-HCl (pH 7.4), and 5 mM MgCl2). Solubilized microsomes were subjected to immunoprecipitations. Nonspecifically bound proteins were removed by preclearing twice with 3 µl of normal rabbit serum or mouse ascites for 45 min with 50 µl of protein G beads (ImmunoPure immobilized protein G beads; Pierce, Rockford, IL). Immunoprecipitations with mAbs Tu36 or Pin-1 were performed for 1 h on a shaker at 4°C. Immune complexes were removed by adsorption onto 50 µl of protein G beads. Pelleted protein G beads were washed four times in wash buffer, containing 0.5% Nonidet P-40, 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, and 5 mM EDTA. Bound immune complexes were eluted in reducing SDS sample buffer (62.5 mM Tris-HCl (pH 6.8), 10% glycerol, 5% 2-ME, and 4% SDS) for 5 min at 95°C and subjected to SDS-PAGE.
Transfection and T cell stimulation assay
DRB1*0301+ DMB wild-type and mutant human
B-LCL (18) were transfected with the Mtv-7 sag
gene under control of the human -actin promoter (15).
Stable transfectants were screened for the level of DRB1*0301
expression by immunofluorescence, and matched clones were used for the
functional stimulation of the V6+ T cell
hybrid, RG17 (19), at various responder to stimulator
ratios. T cell stimulation was measured by the level of IL-2
production, assessed in a bioassay by the proliferation of the
IL-2-dependent cell line HT-2. Transfected cells were cultured in
complete medium at 5 x 105/ml in the
presence of G418 (1 mg/ml). The expression of Mtv-7 sag mRNA
in stable transfectants was monitored by Northern blotting. FACS
analysis with anti-HLA.DR mAb (L243; BD PharMingen, San Diego, CA)
was performed to assay the expression of DRB1*0301 in both the
transfected wild-type (8.1.6) and DMB mutant (9.5.3) cells. The mean
fluorescence intensity of DRB1*0301 expression in each cell line was
between 98 and 102. Stable transfectants were washed and mixed with
2 x 105 V6+ RG17 T
cell hybrids in 0.4 ml of complete medium and incubated for 24 h
at 37°C. Plates were then frozen, and 100 µl of thawed supernatant
was tested in quadruplicate for IL-2 concentration using 2 x
104 HT-2 cells/well in a 96-well round-bottom
plate. Proliferation of HT-2 cells was measured by incorporation of
[3H]thymidine (1 µCi/well). Chinese hamster
ovary cells transfected with murine I-Ek,
CHIEpuro (20) were transfected with
wild-type Mtv-7 sag (CHIE/S7) (20) or the P1S7
mutant gene under control of the human -actin promoter phA-P1S7,
using Lipofectamine (Life Technologies, Gaithersburg, MD). Two days
after transfection, the cells were trypsinized, washed, and plated at
10, 102, or 103 cells/well
in flat-bottom 96-well plates in the presence of G418 (1 mg/ml) and
puromycin (25 µg/ml; Sigma, St. Louis, MO). Stable transfectants were
selected for expression of P1S7 by Northern blotting and for
I-Ek by FACS analysis using mAb 14.4.4s-FITC (BD
PharMingen, San Diego, CA). Clones expressing high levels of both P1S7
and I-Ek were tested for T cell stimulation using
an equal number of the V6+ T cell hybridoma
Omls (20) (2 x 104/well;
stimulator/responder ratio, 1:1). IL-2 production was measured by HT-2
cell proliferation.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
Class II molecules bind peptides generated by proteolysis in the
endocytic pathway (1). Before the class II dimers can be
loaded with peptides, Ii must be proteolytically destroyed, and CLIP
needs to be dislodged from the peptide binding groove. CLIP removal and
subsequent peptide loading are catalyzed by a class II-like molecule,
DM (5). Cell lines deficient in DM have a reduced capacity
to present externally added protein Ags, while no defect is seen in the
presentation of short synthetic peptides (5, 6). We show
here that presentation of the Mtv-7 superantigen by class
II, as measured by T cell recognition, does not require DM. The
DRB1*0301-expressing DMB mutant (9.5.3) and wild-type (8.1.6) B
lymphoblastoid cell lines (18) were transfected with
Mtv-7 sag (15). Stable transfectants, matched
for the level of DRB1*0301 on their cell surface and Mtv-7
sag mRNA (data not shown), were assayed for superantigen
expression by their capacity to stimulate the
V6+ T cell hybrid, RG17 (19). No
difference in stimulatory capacity was seen between DMB wild-type and
mutant cells (Fig. 1
). The phenotype of
the DMB mutant cell line was confirmed by staining with an
anti-CLIP mAb, 1-5, which recognizes class II molecules only from
DM mutants, but not from wild-type cells (data not shown). If
presentation of the Mtv-7 superantigen were due to
proteolytic processing of the Ag in the endocytic pathway, the efficacy
of presentation would be compromised by the absence of DM. Because no
such effect was observed, our results are consistent with the idea that
the Mtv-7 superantigen is presented intact. These results
raise the question of how a protein synthesized by the APC itself can
associate with MHC class II and be presented on the cell surface.
Furthermore, in which cellular compartmentdoes the MMTV superantigen
associate with class II? We postulate that the association of MMTV
superantigen with MHC class II occurs in the same manner as the
interaction of class II with Ii.
|
Both Mtv-7 superantigen and Ii are type II
transmembrane glycoproteins of similar size (Fig. 2A) (16, 21), and
both are known to associate with the MHC class II heterodimer.
The association of Ii and MHC class II involves the class II peptide
binding groove and takes place in the ER via the CLIP region of Ii
(10, 11, 22). We scanned the Mtv-7 superantigen
amino acid sequence for a motif that fulfills the requirements for
class II binding, applying the side chain scoring algorithm developed
by Hammer et al. (17). The peptide score, derived from the
sum of corresponding side chain values of the quantitative motif, has
to be >2 for a peptide to be considered to have binding affinity for
MHC class II. The region spanning aa 86–95 of the Mtv-7
superantigen (YNLNNSENS) exhibits a high class II binding score of 5.8
(Fig. 2A). Interestingly, this CIIPBM is conserved among
various MMTV superantigens and occurs at a location similar to that
of CLIP in intact Ii (Fig. 2A). In comparison, the Ii CLIP
(MRMATPLLM), which has no amino acid identities with CIIPBM defined in
the Mtv-7 superantigen, has a score of 3.9. Similarly, the
influenza virus HA peptide 308–316, which was determined by x-ray
crystallography to bind HLA DR1 in an identical manner to that in which
Ii CLIP binds HLA DR3 (11), was predicted to have a
binding score of 4.3. The presence of a CIIPBM in the Mtv-7
superantigen led us to hypothesize that the MMTV superantigen may be
structurally analogous to Ii and thus associate with MHC class II in a
similar manner.
|
heterodimer in an
in vitro translation system
To investigate possible interaction of the Mtv-7
superantigen and class II molecules, we used an in vitro translation
system, previously shown to support the early steps of class II
assembly (14). Class II dimers, generated by in
vitro translation in rabbit reticulocyte lysate in the presence of
microsomes, are indistinguishable from those found in living cells in
terms of their ability to associate with intact Ii or bind peptide.
Because the in vitro assembly of DR1 dimers has been described
in detail (14), we chose this class II substrate to
explore a possible interaction with the Mtv-7 sag product.
This superantigen is presented equally well by HLA-DR1 and by mouse
class II I-E molecules to murine V6+ T cell
hybrids (23) and human T cells (24). Two
chimeric constructs, IM1 and IM2, consisting of the cytoplasmic and
transmembrane domains of Ii and the extracellular C-terminal domain
of Mtv-7 sag (see Fig. 2
A), were generated for
use by in vitro translation with DR1 - and -chains. These
chimeric molecules can be recognized by the mAb Pin-1, directed against
the N terminus of Ii (25). IM1 includes the N-terminal
portion of Ii, aa 1–106, which contains CLIP, while IM2 has a similar
truncated portion of Ii, aa 1–90, that excludes the CLIP region. In
vitro translation studies of truncated Ii chain with HLA-DR1 and
- have demonstrated that the CLIP region between aa 96 and 104 is
critical for association with class II, as deletion of this region
abolished binding (10).
The cDNAs encoding IM1, IM2, and HLA-DR1 - and -chains were used
for the production of mRNAs that were translated, either separately or
in combination, in the presence of canine pancreatic microsomal
membranes. Immunoprecipitations were performed on detergent extracts
from these microsomes with an mAb against the assembled DR1 complex,
Tu36 (26), or against the cytoplasmic tail of Ii, Pin-1
(25). The IM1 and IM2 translation products can be
recognized by mAb Pin-1, but do not react with either mAb Tu36 or
rabbit anti-HLA DR and anti-HLA DR polyclonal Abs (data
not shown). When mRNAs of IM1 or IM2 were cotranslated with DR1 and
- mRNAs in vitro, immunoprecipitation with Tu36 resulted in recovery
of dimers, as well as /IM1 and /IM2 complexes, but
not IM1 or IM2 alone (Fig. 2B). In reciprocal fashion, Pin-1
immunoprecipitated the IM1 and IM2 translation products as well as
/IM1 and /IM2 complexes, but not dimers alone (Fig. 2B). Thus, IM1 and IM2 can associate with the newly
synthesized DR1 heterodimer. The IM1 construct served as a
positive control in these experiments, because it contains the
Ii-derived CLIP region. The fact that IM2, which lacks CLIP, associated
with DR1 as well as IM1 indicates that CIIPBM in the superantigen may
contribute to the observed binding. IM2 did not associate with either
free - or -chain of class II when cotranslated in vitro. This
suggests that formation of the /IM2 complexes can occur only
through an dimer intermediate (data not shown).
To confirm the role of the class II binding motif in the association of
the MMTV superantigen with class II molecules, we changed the P1 and P2
positions from tyrosine and asparagine to alanine and glycine,
respectively (Fig. 2A), destroying the anchor position of
the peptide binding motif. As predicted, the ability of this mutant,
m86IM2, to associate with the DR1 heterodimers was dramatically
reduced in the in vitro translation reaction (Fig. 2C).
These data suggest that the CIIPBM region involved in the association
between MMTV superantigen and MHC class II is analogous to that of Ii
CLIP with class II. Competitive inhibition studies indicated that Ii
CLIP bound with higher affinity to HLA DR1 than the CIIPBM, while a
negative control peptide had no influence on the interaction between
Mtv-7 and DR1 (data not shown). The C-terminal domain of recombinant
soluble Ii protein forms an -helix that trimerizes and is able to
interact with empty class II molecules in the ER. It has been suggested
that this trimerization augments the affinity of the Ii chain for class
II dimers (27). No trimer or oligomer formation of
the MMTV superantigen could be detected in the in vitro translation
system under conditions in which Ii trimers are readily seen (data not
shown) (10). This lack of trimer formation offers a
possible explanation for findings in Ii knockout mice
(28), in which the MMTV superantigen was unable to
override the Ii defect, as judged by the level of MHC class II
expression and CD4 T cell development. It is possible that the
synthesis rate of the Mtv-7 sag is much slower than that of
Ii in vivo.
Mutation in Mtv-7sag CLIP affects functional superantigen expression
To further confirm the importance of the class II binding motif in
the Mtv-7 superantigen, we introduced the m86 mutation in
the intact Mtv-7 sag gene, referred to as P1S7.
This was subcloned into the eukaryotic expression vector phAPr.1.neo
and transfected into Chinese hamster ovary cells expressing mouse
I-Ek, CHIE cells (20). Stable
transfectants were derived and tested for mutant P1S7 mRNA
by Northern analysis and for levels of class II expression by staining
with the anti-IEk mAb 14.4.4s. Clones with
matched expression (Fig. 3, A
and B) were tested for superantigen expression by
stimulation of the V6+ T cell hybrid, Omls
(20), compared with wild-type Mtv-7
sag-transfected CHIE cells, CHIE/S7 (20) (Fig. 3C). T cell stimulation was measured by the level of IL-2
production, as detected by proliferation of the IL-2-dependent cell
line, HT-2. The mutation in the Mtv-7 sag CIIPBM
negatively affected functional presentation of superantigen to T cells
(Fig. 3C), suggesting that the mutation inhibits
superantigen presentation at the cell surface.
|
Previous biochemical studies by Winslow et al. have shown that the
Mtv-7 superantigen is synthesized in the ER as a 45-kDa
glycoprotein (gp45), while an 18-kDa carboxyl-terminal fragment is
detected at the cell surface (29). Because gp45 is by far
the predominant form detected in immunoblot analyses, most
Mtv-7 superantigen appears to be retained and degraded in
the ER. It is presently unclear how the MMTV superantigen leaves the
ER, but it is possible that the limiting factor is the inefficiency
with which MMTV competes with Ii for binding to the MHC class II
heterodimer during assembly. MMTV superantigen is not detected on the
surface of B lymphocytes of inappropriate MHC class II haplotypes, as
assessed by immunochemical and immunofluorescent techniques
(29), implying that MMTV superantigen presentation in vivo
requires the appropriately matched class II product. Because mice that
lack Ii show reduced level of MMTV superantigen activity
(28), Ii could play a role in the trafficking of the MMTV
superantigen/class II complex, perhaps by forming mixed trimers,
composed of Ii and superantigen. Mixed oligomers of the p31 and p41
isoforms of Ii, the latter a splice variant of Ii that contains an
additional 80 residues, have been described and demonstrate the ability
of Ii to engage in mixed oligomer formation (30). Such
mixed complexes may be able to exit the ER and follow the MHC class II
maturation pathway through Golgi/post-Golgi and endocytic compartments
(Fig. 4). The trans-Golgi
network is the site where furin, a protease previously implicated in
selective proteolytic processing of the Mtv-7 superantigen
(29), is located. Based on the intracellular distribution
of furin, its action precedes that of the proteases in the endocytic
pathway, which might also act on the MMTV superantigen (see model in
Fig. 4). This hypothesis fits with the observation that no full-length
membrane form of the MMTV superantigen has been detected at the cell
surface, whereas an 18-kDa carboxyl-terminal fragment remains
associated with the class II protein on the cell surface
(29). Thus, the CIIPBM might be necessary, but not
sufficient, for binding, because the 18-kDa fragment might presumably
also contribute to binding.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Graduate Institute of Immunology, College of Medicine, National Taiwan University, Taipei, Taiwan.
3 Current address: Department of Microbiology, Ohio State University, Columbus, OH 43210.
4 Address correspondence and reprint requests to Dr. Brigitte T. Huber, Department of Pathology, Tufts University School of Medicine, 136 Harrison Avenue, Boston, MA 02111.
5 Abbreviations used in this paper: MMTV, mouse mammary tumor virus; ER, endoplasmic reticulum; Ii, invariant chain; CLIP, class II-associated Ii peptide; CIIPBM; class II peptide binding motif; P1, position 1.
6 Abbreviations used in this paper: MMTV, mouse mammary tumor virus; ER, endoplasmic reticulum; Ii, invariant chain; CLIP, class II-associated Ii peptide; CIIPBM, class II peptide binding motif; P1, position 1.
Received for publication November 2, 2000. Accepted for publication December 18, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResults and DiscussionReferences |
|---|
chain of HLA-DR transplantation antigens is a member of the same protein superfamily as the immunoglobulins. Cell 30:153.[Medline]
chains from DR1 haplotype. Proc. Natl. Acad. Sci. USA 82:3405. restriction and conserved evolutionary features. J. Exp. Med. 177:1735.This article has been cited by other articles:
|
|
 |
|
  G. A. Azar, R.-P. Sekaly, and J. Thibodeau A Defective Viral Superantigen-Presenting Phenotype in HLA-DR Transfectants Is Corrected by CIITA J. Immunol., June 15, 2005; 174(12): 7548 - 7557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Neumann, A. M. Eis-Hubinger, and N. Koch Herpes Simplex Virus Type 1 Targets the MHC Class II Processing Pathway for Immune Evasion J. Immunol., September 15, 2003; 171(6): 3075 - 3083. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wirth, A. Vessaz, C. Krummenacher, F. Baribaud, H. Acha-Orbea, and H. Diggelmann Regions of Mouse Mammary Tumor Virus Superantigen Involved in Interaction with the Major Histocompatibility Complex Class II I-A Molecule J. Virol., October 2, 2002; 76(21): 11172 - 11175. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
