| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,,
*
Laboratoire d’Immunologie, Département de Microbiologie et Immunologie, Université de Montréal, Montréal, Canada;
Department of Experimental Medicine, McGill School of Medicine, Montréal, Canada;
Laboratoire d’Immunologie, Centre de Recherche du Centre Hospitalier de l’Université de Montréal, Montréal, Canada;
Department of Microbiology and Immunology, McGill University, Montréal, Canada;
¶ Institut Pasteur, Unité d’Immunochimie Analytique, Paris, France; and
||
Laboratoire d’Immunologie Moléculaire, Université de Montréal, Montréal, Canada
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain variable domain (4). T cell
stimulation by SAgs leads to suppression of IL-2 production, anergy,
and apoptosis of SAg-reactive T cells (5, 6). The
determinants responsible for the induction of T cell activation and/or
deletion by SAgs remain enigmatic and thus far cannot be explained by
their kinetics and/or topology of interaction with the TCR
(7). Indeed, it has been demonstrated by surface plasmon
resonance that the affinity of SAg-class II complexes for soluble TCRs
is well within the range defined for agonist peptide ligands (8, 9). A better understanding of the molecular parameters allowing
SAgs to induce a tolerance state in the host should allow a more
rational use of these toxins in the development of immunomodulatory
therapies. Bacterial SAgs are presented in their native form and bind conserved regions of MHC class II molecules outside the peptide-binding groove (10). As such, they should bind to the totality of cell surface MHC class II molecules and thus engage the TCR with a higher avidity than nominal Ags. However, we and others have shown that SAg binding and presentation is markedly affected by the structural microheterogeneity of the MHC class II molecules, probably imparted by the groove-occupying peptide (11, 12, 13, 14, 15, 16). Whether that phenomenon modulates the activity of SAgs on T cells has not been addressed in previous studies. In vivo, the proportion of MHC molecules bound by a given antigenic peptide appears to be naturally limited by inherent peptide processing or loading mechanisms. This possibly prevents induction of high avidity T cell tolerance against nonself Ags (17). Similarly, the restriction imposed by the groove peptide on the binding of SAgs to class II molecules might limit functional inactivation of the responding T cells at low doses (16, 18). Moreover, given that the TCR machinery is sensitive to very low amounts of antigenic peptide ligands (19, 20) or SAg (21), others have proposed that a limited number of SAg-MHC class II complexes is necessary to insure proper serial triggering of the TCR (16).
In the present study, we investigated quantitatively how the structural microheterogeneity in MHC class II molecules affects the binding of staphylococcal enterotoxin A (SEA) to B lymphocytes. We have determined 1) the proportion of MHC class II-peptide complexes on B cells that allows binding of SEA, 2) the minimal number of complexes required for T cell activation, and 3) the number of SAg-MHC complexes required to induce T cell death. Our results show that the amount of SEA receptors on APCs largely exceeds the proportion of class II-ligand complexes required for T cell activation or deletion.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human EBV-transformed B cell line LG2 (DR1/DR1) was provided
by L. J. Stern (Massachusetts Institute of Technology, Cambridge,
MA) and grown in RPMI 1640 (Life Technologies, Rockville, MD)
supplemented with 5–10% FCS and 2-ME. The pan anti-DR Ab L243
(IgG2a) (HB55; American Type Culture Collection, Manassas, VA)
recognizes a conformational epitope situated in the 
1 domain
(22). The anti-DR XD5.117 (23) and
anti-DR DA6.147 (24) mAbs have been previously
described. In all the biochemical experiments, SEA from three
different sources was used and yielded similar results. Purified SEA
was obtained from Toxin Technology (Sarasota, FL), and purified
recombinant SEA was a kind gift of P. Antonsson (Active Biotech
Research, Lund, Sweden) and W. Mourad (Université Laval,
Québec, Canada). The L243 hybridoma was grown in RPMI 1640 medium
supplemented with synthetic 2% HY (Life Technologies) and
purified on a column of protein G-Sepharose 4 Fast Flow (Amersham
Pharmacia Biotech, Quebec, Canada). The Fab portion (L243-Fab) was
prepared using a commercially available kit (Pierce, Rockford, IL) and
yielded a single band (50 kDa) on SDS-PAGE under nonreducing
conditions. The concentration was determined by absorbance at 280 nm on
a standard spectrophotometer. The SEA or L243-Fab (referred to as
SEA-bio or L243-bio) was biotinylated using the
sulfo-NHS-LC-biotin as described by the manufacturer (Pierce).
The biotinylated proteins were dialyzed in PBS and stored at -20°C
in 50% glycerol. Sepharose-SEA or -L243 was prepared by coupling
proteins to cyanogen bromide-activated Sepharose 4B beads
(Pharmacia Biotech, Uppsala, Sweden) according to the supplier’s
recommendations. In brief, SEA or L243 (600 µg) in 0.1 M
NaHCO3 and 0.5 M NaCl was incubated to cyanogen
bromide-activated Sepharose gel (0.75 g) for 2 h at 4°C with
constant agitation. The beads were then blocked with ethanolamine,
washed, and kept at 4°C.
Purification of DR1 molecules
Soluble HLA-DR1 was produced in a baculovirus-based system (from the pACDR1 virus obtained from L. J. Stern) and purified according to protocols described (25). Purified membrane HLA-DR1 was obtained as follows: Cells (2 x 108) were lysed in 20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10 µM ZnSO4, 1 mM PefaBloc SC (Roche, Montréal, Canada), 10 µg/ml leupeptin, 10 µg/ml aprotinin, and 1% Triton X-100, and the insoluble materiel was spun at 14,000 rpm for 20 min. The supernatant was passed over a L243 column and eluted (see above). The concentration of the DR1 preparation was determined immediately by the Bradford method, and Triton was added to a final concentration of 1%. The HLA-DR1 heterodimer migrated as a homogeneous compact band (55 kDa) in SDS-PAGE under nonboiled conditions.
Surface binding experiments and Scatchard analyses
A dozen completely independent experiments were performed over a several-month period. The concentration of each batch of purified SEA or L243-Fab was carefully determined on Coomassie-stained SDS-PAGE and by Bradford protein dosage (data not shown). The radiolabeling and fractionation of SEA or L243 was performed as described (12). SDS-PAGE separation of the labeled fraction confirmed that the ligand of interest was responsible for >95% of the total specific activity. The binding assay was performed by incubating 125I-radiolabeled SEA or L243-Fab (25–100 ng, depending on the experiments) with 200,000–106 LG2 cells (varied between the different experiments) for 4 h in DMEM (2% FCS-0.05% NaN3) in the presence or absence of a 0.1- to 100-fold excess of unlabeled competitor. The bound fraction was separated from free ligand by centrifugation of cells over a silicon oil cushion (16% mineral oil and 84% silicon oil; Sigma, St. Louis, MO) and counted.
The free fraction of ligand (F), as well as the total
(nonspecific and specific) amount of ligand bound
(Btot) at a given concentration, was
calculated using the following equations:
 |
 |
when
[F] >> Kd (i.e., when the specific
binding is saturated); L signifies the amount of cold ligand
added; L* signifies the amount of
125I-labeled ligand added; [F]
signifies the concentration of free ligand in solution;
Kd signifies equilibrium dissociation
constant of the interaction; and Bmax
signifies the number of receptor molecules. Human PBL purification, cytokine production, and [3H]thymidine incorporation assays
Peripheral blood obtained from a DR1 donor was diluted (2/1) in
PBS and underlayered with Ficoll-Hypaque (Pharmacia Biotech) at room
temperature. After centrifugation, the interface was collected and
washed three to four times in PBS-2% FCS. MHC class II-negative PBLs
were obtained using a modified protocol of the goat
anti-mouse-conjugated IsoCell enrichment columns (Pierce).
Cells were incubated for 45 min with anti-CD14 (Leu-14), -CD16
(Leu-16), -CD19 (Leu-19), -CD56 (Leu-56) (BD Biosciences, San Jose CA),
and L243 Abs at 4°C and passed through the enrichment column. Cell
fractions were collected and washed in RPMI 1640–10% FCS. The
quality of the purification was confirmed by staining for HLA-DR (using
the 50D6 mAb) anti-TCR (Beckman Coulter, Miami FL), and CD4
and CD8 (Simultest; BD Biosciences). For
[3H]thymidine incorporation assays, cells were
cultured at 37°C in complete RPMI 1640 medium supplemented with 5%
FCS. Total PBMCs alone or PBLs with LG2 cells were incubated in the
presence of SEA in round-bottom 96-well plates. Following incubation, 1
µCi [3H]thymidine was added for 16 h at
37°C. Cells were harvested, and [3H]thymidine
incorporation was measured using a beta plate counter (Pharmacia LKB,
Gaithersburg, MD). For [3H]thymidine
incorporation assays using purified T cells, LG2 cells were washed
three to four times in PBS, treated for 1.5 h with 100 µg/ml
mitomycin C (Sigma) at a density of 107 cells/ml,
and washed again three to four times in PBS 2%-FCS to block DNA
synthesis. IL-2 production was assayed by incubating SEA with total
PBMCs (105) for 24 or 36 h. IL-2 in the
supernatant was quantified using a CTLL-based assay as previously
described (11). Intracellular IL-2 and IFN-
were
measured after 8 h of stimulation using procedures previously
described (26). In brief, PBMCs (2 x
106/ml) were incubated with no stimulus, PMA (25
ng/ml) and ionomycin (1 µg/ml), or SEA for 8 h. Brefeldin A (10
µg/ml) was added for the final 6 h of stimulation. After
stimulation, cells were permeabilized and stained with either
anti-IL-2 or anti-IFN-, anti-CD69, and either
anti-CD4 or anti-TCR. Abs, isotype controls, and lysing and
permeabilizing solutions were purchased from BD Biosciences. Control Ab
for permeabilization was purchased from Medicorp (Montreal, Quebec,
Canada). Flow cytometry analyses were performed on a FACScan using
CellQuest software (BD Biosciences). In lymphocyte stimulation
experiments, recombinant SEA was used.
Quantitative immunoprecipitations and Western blotting
Cell surface HLA-DR1 molecules were immunoprecipitated using SEA-bio or L243-bio. Unless specified otherwise, LG2 cells (3 x 105) were incubated for 4 h in presence of either SEA-bio (80 µg/ml; 3 µM), or L243-bio (60 µg/ml; 0.6 µM) at 37 or 4°C in culture medium stabilized with 0.01% sodium azide. After incubation, the cells were washed eight times in 1 ml of PBS to eliminate background, as controlled using an irrelevant biotinylated mAb. Cells were then lysed in 200 µl lysis buffer and kept on ice for 15 min. The insoluble material was spun down, and the supernatant was transferred to a fresh Eppendorf tube containing 100 µl of streptavidin-coated nonporous beads (Dynal M-280; Dynal Biotech, Oslo, Norway). After 1 h at 4°C, 1 ml of PBS was added, and the beads were centrifuged for 10 s. The beads were again washed 10 times (lysis buffer with 0.5% Triton X-100) and resuspended in 50 µl of 2x Laemmli protein-loading buffer. The samples were boiled for 5 min, loaded on 10% SDS-PAGE, and analyzed by Western blotting using the XD5.117 and/or DA6.147 Abs as previously described (12). A peroxydase-coupled goat anti-mouse secondary Ab reactive to the Fc portion of mouse Igs was used for chemiluminescence detection (Pierce).
Depletion experiments
LG2 cells (2 x 105) were incubated as indicated above in 50 µg/ml SEA-bio for 4 h at 37°C before washings and immunoprecipitation with streptavidin-coupled beads. The lysate was transferred to an Eppendorf tube containing fresh SEA-bio-streptavidin beads. After a 4-h agitation at 4°C, the beads were centrifuged, washed eight times with lysis buffer (20 mM Tris-HCl (pH 7.5), 150 mM NaCl, 10 µM ZnSO4, 1 mM PefaBloc SC, 10 µg/ml leupeptin, 10 µg/ml aprotinin, and 1% Triton X-100), and resuspended in 50 µl of 2x Laemlli buffer. The lysate was then reincubated for three more rounds of depletion using saturating amounts of SEA-coupled Sepharose beads (400 µg/ml SEA). Alternatively, similar results were obtained (data not shown) by extraction of the lysate for 4 h at 4°C in the presence of SEA-bio and immunoprecipitation by transferring the lysate to a fresh Eppendorf tube containing a pellet of 3 x 108 streptavidin beads. The lysate was then transferred to a fresh tube containing 75 µl of L243-coupled Sepharose beads and incubated at 4°C for an additional 4 h before washings. The samples were loaded on SDS-PAGE and analyzed by Western blotting.
CFSE, annexin V, and 7-amino actinomycin D (7-AAD) experiments
The CFSE dye was obtained from Molecular Probes (Eugene, OR).
CFSE (0.5–1.25 µM, depending on the preparation) was added to 2
x 107 PBMC and incubated with gentle mixing at
room temperature for 10 min. The reaction was quenched by addition of
an equal volume of FCS, and cells were washed three times with PBS
containing 5% FCS. The cells were cultured at 37°C at 5%
CO2 at 1.5 x 106/ml
in RPMI 1640 containing 3% FCS overnight to obtain a stable
fluorescence intensity between 103 and
104 logs. Labeling of cells with 7-AAD and
annexin was performed as previously described (27). For
the experiment in Fig. 7
, cells stained with CFSE according to the
above-mentioned method were stimulated with SEA or PHA for 3 days and
then labeled with 7-AAD before analysis by flow cytometry.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Scatchard analyses.
We have previously shown that B cell lines or primary lymphoid cells
expressing the class II-associated invariant chain and HLA-DM can bind
and present SAgs most efficiently (12). To determine the
fraction of MHC class II molecules that are permissive to SEA binding,
we performed Scatchard analyses using radiolabeled SEA. For this
purpose, we used the representative DR1-expressing B cell line LG2
because, in those cells, the great majority of MHC class II molecules
form SDS-stable complexes containing a mixture of oligopeptides
(25). Although B lymphocytes express other class II
isotypes, HLA-DR is the major receptor for this SAg (28).
The reproducibility of the number of complexes obtained by Scatchard
plots is shown by SDs of <2% obtained between each of the independent
experiments (see Table I). A
representative set of data is shown in Fig. 1A. Of note, in each of these experiments,
the binding of either SEA (n = 7) or the L243
anti-DR Fab (n = 4) appeared to follow a
first-order binding ligand-receptor interaction as previously described
by others (29, 30). Binding of SEA occurs mostly to the
high affinity -chain site, as occupancy of the low affinity of the
-chain site (Kd 
10-5 M) is insignificant in the range of
concentrations used (29, 30, 31, 32). Using these data, we
calculated that LG2 cells express a total of (3.2 ± 0.5) x
106 SEA receptors/cell. An average
Kd of (4 ± 2) x
10-8 M was derived, which is in accordance with
affinities reported on a number of cell types (12, 33).
Similarly, we found a total number of (5.2 ± 2.6) x
106 DR1 receptors/cell as measured using an
anti-DR Fab portion (Table I). These experiments show that the
proportion of SAg-binding sites averages 60% of the total number of
surface MHC class II molecules, suggesting that some class II molecules
on the surface of APCs are unable to interact with SEA.
|
|
), we used
independent approaches to confirm the number of DR1 molecules expressed
at the surface of LG2 cells. As shown in Fig. 1B, the amount
of cell surface DR1 molecules immunoprecipitated using saturating
concentrations of the L243 Fab portion was compared with DR1 standards
of known concentrations. In at least five independent experiments, the
total number of HLA-DR1 receptors found ((5.3 ± 0.6) x
106 molecules/cell) was highly consistent with
results obtained on Scatchard plots.
The heterogeneity of antigenic peptides bound to MHC class II molecules
at the cell surface may give rise to interactions with SEA of different
affinities (13), thus skewing our estimate of the total
number of binding sites derived from Scatchard data. In principle, such
lower-affinity sites can be discerned from nonbinding sites by using
saturating amounts of the ligand. We have confirmed the binding
of SEA to a fraction of class II molecules by quantitatively
comparing the total surface HLA-DR molecules immunoprecipitated
using microgram amounts of L243 vs SEA. The amount of DR1 was
determined by Western blotting using the XDS.117 Ab, which recognizes a
linear epitope on the DR -chain (34). Fig. 2A shows that about half
(49 ± 8.3%) of MHC class II molecules were immunoprecipitated
with SEA as compared with L243. Similar results were obtained in five
different experiments independently of the temperature used for
incubation, and these are in accordance with Scatchard results (see
Table I). Importantly, the nonconformer anti-DRb Ab XD5.117
precipitated an amount of DR1 comparable to that obtained with L243,
indicating that the latter does not bind to a limited subset of MHC
class II in itself (data not shown).
|
A), followed by a final
immunoprecipitation using the anti-DR Ab L243. We specifically
chose to use this conformation-dependent mAb, which recognizes only
fully folded post-Golgi heterodimers of importance here
(35). Fig. 2B is a representative experiment in
which we completely depleted SEA-binding HLA-DR receptors from
the LG2 lysate after three to four rounds of
immunoprecipitation. However, it was not possible to deplete all of the
HLA-DR molecules from the cell extract, as demonstrated by the presence
of a strong residual signal for DR molecules immunoprecipitated using
L243. Altogether, the results obtained from Scatchard plots,
immunoprecipitation, and depletion experiments provide a rigorous
quantification of the number of MHC class II-peptide complexes that are
structurally competent to bind SEA. These receptors represent
45–60% of the total surface HLA-DR molecules, thus suggesting that
some class II-bound oligopeptides might restrict the binding of SAgs on
live cells. The number of cell surface MHC class II-SAg complexes required for T cell stimulation
LG2 cells were used in proliferation assays to characterize the
relationship between the number of class II-SAg complexes and the
extent of the T cell response. Fig. 3A compares a typical
proliferation dose-response curve to the proportion of HLA-DR1
molecules that are occupied by SEA at different concentrations.
Independently of the T cell:LG2 ratio, proliferation was detected
starting at concentrations as low as 10-14 M SEA
(Fig. 3A), as previously described (36).
Similarly, blastic transformation was evident by flow cytometry at
10-15 M SEA (Fig. 3B). These
concentrations correspond to an average SEA occupancy of 0.1
(0.000003125%) to 1 (0.00003125%) DR1 molecules/cell (Fig. 3C). Most likely, at these concentrations, only some APCs
are occupied by a few SEA ligands per cell and are therefore able to
stimulate a small number of T cells, possibly those of highest affinity
for the SAg-class II complex. The potency of SAgs such as SEA is
remarkable because an extremely low number of SEA-MHC class II
complexes are sufficient to trigger T cell activation in the conditions
established in the present assay. Many more potential SAg-MHC class II
complexes can be formed than the minimal number required for T cell
activation (see Fig. 3A). Maximal T cell proliferation was
achieved at SEA concentrations corresponding to 10,000–20,000
occupied MHC class II molecules (0.2 nM SEA).
|
A), suggesting that
a higher density of SAg ligands leads to an impaired T cell response.
This decrease in proliferation could not be explained by a lower
efficiency of TCR engagement, as evidenced by the strong
internalization of the TCR in SEA-reactive
V22+ T cells (which account for a large part
of the SEA-responsive T cells; data not shown) above 0.2 nM SEA. As
shown in Fig. 3D, TCR internalization was rapid and occurred
within the first hour of stimulation. The level of TCR rapidly
decreased by 10% within an hour under conditions in which 10
SAg-MHC class II complexes were formed. At saturating doses of SEA
(10-7 M or 2 x
106 SEA-MHC complexes/cell), >90% of TCRs were
internalized from the cell surface. As a control, TCR internalization
was not observed in the non-SEA-responsive TCR V2 cells (data not
shown). Presentation of supraoptimal number of SEA-MHC class II complexes leads to impaired cytokine production by T cells
The next series of experiments were designed to study the
functional response of T cells to supraoptimal doses of SEA as defined
above. The decreased T cell proliferation observed after 3 days at high
SEA concentrations (Fig. 3A) was not exclusive to
presentation by LG2 cells, as the same results were obtained using
class II-autologous PBLs as a source of APCs (Fig. 4A). At supraoptimal SEA
concentrations, T cells underwent a normal dose-dependent kinetics of
induction of early (3–18 h) CD69 and late (24–48 h) CD25 activation
markers (see Fig. 4, B and C). However, above
10-10 M, a selective dose-dependent inhibition
of IL-2 production was observed (Fig. 5A). For example, at
10-7 M SEA (Fig. 5B), we measured a
2-fold reduction in the proportion of cytokine (IL-2 and
IFN-)-producing cells among activated (CD69+)
lymphocytes (Fig. 5C) as early as 8 h after
stimulation.
|
|
Despite the normal induction of CD69 and CD25, the proliferative
response of T cells at high doses of SEA dropped after 3 days (see
Figs. 3 and 4, A). At day 5, the number of cells recovered
at 10-7 M SEA was barely increased over the
unstimulated control (Fig. 6A). In contrast, at the
optimal dose of 10-11 M SEA, considering that
SEA triggers only 10–15% of total population of T cells, a 7-fold
expansion was observed. Importantly, the reduction in expansion at high
SEA doses was not due to nutrient depletion following a massive
expansion, because stimulation with PHA, which roughly triggers
proliferation of 10-fold more T cells, yielded a significant expansion
(see Fig. 6A).
|
B, the proportion of nonviable
(7-AAD+) cells after 48 h was significantly
increased to >50% as cells were stimulated by supraoptimal
(10-5 M) doses of SEA. The commitment of
activated T cells to apoptosis was further confirmed by the presence of
annexin V+7-AAD- cells (a
phenotype associated with early apoptosis (27)) in the
cells stimulated by SEA.
In additional experiments, we used the CFSE dye that specifically
tracks dividing cells to confirm the relationship between the
proliferation of SEA-responding cells and their commitment to
apoptosis. As described previously, CFSE is a marker of cell division
because the fluorescent dye is halved in each daughter cell
(37). A representative experiment is shown in Fig. 7. Activated T cells expanded under all
three conditions tested, consistent with our observation that the
induction of activation markers is unaffected by supraoptimal levels of
SEA in the first 48 h after stimulation (Fig. 4, A and
B). Interestingly, we observed a late occurrence of cell
death (as measured by staining with 7-AAD) specifically after three to
four rounds of division in cells stimulated with SEA. This result is
consistent with the absolute requirement for T cell proliferation in
SAg-induced apoptosis in vivo (38). Although we detected
some degree of cell death in all conditions tested, apoptosis increased
in a dose-dependent manner at supraoptimal vs optimal SEA doses (53 vs
13%) (see Fig. 7).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The inability of SAgs to bind all the cell surface HLA-DR molecules is a likely a consequence of their indispensable need to interact in close proximity to the polymorphic and structurally diverse peptide groove to reach the TCR (41). Although our results demonstrate that a majority of peptides are permissive to SEA binding, we cannot formally exclude that variations in the affinity of different MHC-peptide complexes may be important for T cell activation. Indeed, among the 35–60% of surface DR1s that bind SEA, ligand-induced structural variations may give rise to a spectrum of affinities for the SAg (13). If the intrinsic affinity of MHC molecules is high (i.e., >4 x 10-8 M), then the existence of progressively lower affinity sites should keep the amount of bound SAg within a range optimal for T cell proliferation. This would be consistent with the idea that a mechanism to limit SAg density is important for inducing a proliferative T cell response as opposed to deletion (16). However, the affinity of SEA as measured herein (4 x 10-8 M) is identical with the highest affinities reported using homogenous soluble MHC class II molecules covalently attached to single peptides. This suggests that, for those MHC complexes that are permissive to SEA binding, the mixture of peptides does not significantly modulate the affinity for the SAg.
The structural difference between SEA-binding vs non-SEA-binding subsets of MHC class II molecules remains to be determined. In addition to MHC class II-bound oligopeptides, other glycoproteins associated in complex with HLA-DR molecules may directly interfere with the binding of SAgs (12, 42). For example, HLA-DR was found to be associated in B cells with molecules such as CD40 (43), HLA-DM (44), and members of the tetraspan protein family (45).
A low amount of SEA, corresponding to 15,000 ligands/APC (0.3% of
total cell surface HLA-DR molecules), triggers a strong proliferative
response with very little apoptosis. Interestingly, in vivo studies
looking at the induction of peripheral deletion by SAgs have used
relatively (1–50 µg/mouse) high doses (6, 46, 47), whereas studies using low doses of SAgs showed that they
could stimulate T cell expansion just like antigenic peptides
(48, 49). Above a given threshold of SEA-MHC complexes,
this balance is tipped toward induction of apoptosis. These
observations suggest that the potential for a large pool of MHC-bound
SEA is essential to the induction of peripheral tolerance in vivo. In
the physiological setting, T cells typically respond to Ags presented
by 102–104 molecules of
the restricting allele occupied by a given peptide (50).
Interestingly, these estimates compare well to our calculation of the
number of SAg ligands needed to achieve a sustained T cell expansion.
This interpretation is consistent with recent experiments showing that
the affinity of MHC class II-bound SAgs for the TCR is within the range
of purely agonist antigenic ligands (8, 9).
In vivo treatment with high doses of SEA leads to a selective
suppression of IL-2 production that occurs before and independently of
T cell deletion (6, 46). In addition to IL-2, we observed
a selective block in the production of IFN- that seemingly occurred
before the induction of apoptosis in these cells. Indeed, at the
earliest time points of TCR engagement (between 6 and 8 h) and in
the absence of any detectable cells undergoing apoptosis, we observed a
drop in the IL-2 produced, despite a normal induction of the CD69
activation marker (Figs. 2 and 4). Interestingly, a comparable
dose-dependent inhibition of the number of cytokine-producing cells has
been reported in specific CD8+ T cells responding
to supraoptimal lymphocytic choriomeningitis viral loads
(51). It is tempting to speculate that prolonged,
sustained engagement of the TCR by SAgs limits the cytokine response of
T cells perhaps by virtue of the excess number of TCR down-regulated.
In our system, addition of exogenous rIL-2 (50, 100, or up to 500 U/ml)
at the beginning or daily in the culture did not rescue these cells
from programmed death (data not shown). Finally, apoptosis occurred as
a relatively late event after initiation of cell division. Based on the
level of CFSE in the apoptotic cells, we estimate that these cells die
only after having gone through about three to four cell divisions.
Interestingly, a similar observation has recently been reported using
staphylococcal enterotoxin B (52).
Primed or naive T cells may likely differ in their sensitivity to
ligand-induced apoptosis (53, 54). Here, a minimal
fraction of the PBLs was composed of primed T cells before
stimulation with SEA, as evidenced by the expression of either CD69 or
CD25 in <1% of these cells (data not shown). Our data mostly reflect
the combined response of naive T cells differing with respect to their
affinity for SEA. However, it may be interesting to investigate the
functional difference existing between memory and naive T cells or
cells otherwise differing in their TCR V expression.
In conclusion, our results show that despite the existence of different pools of class II molecules relative to their ability to bind SEA, the potential number of SAg-MHC class II complexes largely exceeds the prerequisite for inducing peripheral tolerance in the host. This data should help define the optimal dosage for SAgs in anti-tumor therapies (55)
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Jacques Thibodeau, Laboratoire d’Immunologie Moléculaire, Département de Microbiologie et Immunologie, Faculté de Médecine, Université de Montréal, C.P. 6128, Succursale Centre-Ville, Montréal, Canada H3C 3J7. E-mail address: jacques.thibodeau{at}UMontreal.CA
3 Abbreviations used in this paper: SAg, superantigen; SEA, staphylococcal enterotoxin A; 7-AAD, 7-amino actinomycin D; FSC, forward light scatter.
Received for publication January 30, 2001. Accepted for publication April 12, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain complexed with a superantigen. Nature 384:188.[Medline]
8+ T cells by staphylococcal enterotoxin B. J. Exp. Med. 172:1091. chain in stabilizing TCR-superantigen-MHC class II complexes. Immunity 10:473.[Medline]
chain residues critical for binding of the toxic shock syndrome toxin superantigen. J. Exp. Med. 176:1779. chains by two-dimensional electrophoresis and Western blotting. Hum. Immunol. 10:177.[Medline]
heterodimers in the absence of antigenic peptide. Cell 68:465.[Medline]
chain. J. Immunol. 158:3698.[Abstract]
-helices along the major histocompatibility complex binding cleft are required for staphylococcal enterotoxin A function. Proc. Natl. Acad. Sci. USA 88:7228. monoclonal antibodies bind to an intracellular C-terminal epitope. Hum. Immunol. 53:34.[Medline]
8+, CD4+ T cells in Staphylococcus enterotoxin B-primed mice. J. Exp. Med. 172:1065.This article has been cited by other articles:
|
|
 |
|
  A. K. M. N.-u. Rahman, C. A. Herfst, B. Moza, S. R. Shames, L. A. Chau, C. Bueno, J. Madrenas, E. J. Sundberg, and J. K. McCormick Molecular Basis of TCR Selectivity, Cross-Reactivity, and Allelic Discrimination by a Bacterial Superantigen: Integrative Functional and Energetic Mapping of the SpeC-Vbeta2.1 Molecular Interface J. Immunol., December 15, 2006; 177(12): 8595 - 8603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Llewelyn, S. Sriskandan, N. Terrazzini, J. Cohen, and D. M. Altmann The TCR V{beta} signature of bacterial superantigens spreads with stimulus strength Int. Immunol., October 1, 2006; 18(10): 1433 - 1441. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Lavoie, A. R. Dumont, H. McGrath, A.-E. Kernaleguen, and R.-P. Sekaly Delayed expansion of a restricted T cell repertoire by low-density TCR ligands Int. Immunol., July 1, 2005; 17(7): 931 - 941. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Papageorgiou, M. D. Baker, J. D. McLeod, S. K. Goda, C. N. Manzotti, D. M. Sansom, H. S. Tranter, and K. R. Acharya Identification of a Secondary Zinc-binding Site in Staphylococcal Enterotoxin C2: IMPLICATIONS FOR SUPERANTIGEN RECOGNITION J. Biol. Chem., January 9, 2004; 279(2): 1297 - 1303. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-A. Langlois, Y. E. Fakhry, and W. Mourad Zinc-binding Sites in the N Terminus of Mycoplasma arthritidis-derived Mitogen Permit the Dimer Formation Required for High Affinity Binding to HLA-DR and for T Cell Activation J. Biol. Chem., June 13, 2003; 278(25): 22309 - 22315. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Rajagopalan, M. K. Smart, C. J. Krco, and C. S. David Expression and Function of Transgenic HLA-DQ Molecules and Lymphocyte Development in Mice Lacking Invariant Chain J. Immunol., August 15, 2002; 169(4): 1774 - 1783. [Abstract] [Full Text] [PDF]
|
|
| ||||
) or -SEA (
) is shown in two graphical
representations. In the two bottom panels, the curves
represent the results of nonlinear regressions (y =
[Bmax x
x]/[x +
Kd]) performed on the whole set of binding
data compiled from multiple independent experiments using the same LG2
cells. B, The amount of DR1 at the surface of LG2 cells
was confirmed by immunoprecipitation using saturating concentration (60
µg/ml) of a L243 Fab portion and compared with soluble DR1 standards
purified by immunoaffinity from baculovirus-infected insect
cells (
), 3 x
104 (
