| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Immunology, Mayo Clinic, Rochester, MN 55905
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
8 family in
DQ8.Ii-/- mice was comparable with that of
DQ8.Ii+/+ mice. Systemic IFN-
production following in
vivo challenge with SEB was reduced in DQ8.Ii-/- mice and
were also protected from SEB-induced toxic shock. Although the T cell
response to a known peptide Ag was diminished in
DQ8.Ii-/- mice, DQ8.Ii-/- APCs were capable
of presenting that peptide to primed T cells from wild-type DQ8 mice as
well as to a specific T cell hybridoma. Differentiation of mature B
cells was also affected to a certain extent in DQ8.Ii-/-
mice. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
heterodimers of MHC class II molecules (Ii) in the endoplasmic
reticulum (ER) through distinct contact regions (9). Due
to the presence of trimerization region in the Ii, three molecules of
Ii trimers form a supramolecular complex of a nonamer,
(Ii)3 (6, 10). Aided by the
presence of targeting sequence in the Ii, this complex is then
transported to the endocytic vesicles (11) where selective
Ii degradation occurs leaving a small fragment of the Ii molecule
called class II-associated invariant peptide (CLIP), still attached to
the peptide-binding groove of the MHC class II molecule. Subsequently,
the CLIP is replaced by the peptide fragment generated from the
processed Ag and this process is facilitated by another nonclassical
MHC class II molecule H-2M (in mice) or HLA-DM (in humans)
(12). Thus, Ii performs three important functions in the
expression of functional class II molecules: 1) It helps in proper
folding/assembly of MHC class II heterodimers. 2) It transports
the newly synthesized MHC class II molecules to the endocytic
compartments for peptide loading (3, 4) and 3) most
importantly it occupies the peptide-binding groove on the MHC class II
molecules during their transit from ER to Ag-loading compartments,
thereby preventing the binding of nonimmunologically relevant peptides
in the ER (13). Genetically engineered mice lacking Ii have tremendously helped in elucidating its functions in MHC class II expression and Ag presentation (4, 14, 15, 16, 17, 18, 19, 20, 21, 22). The phenotype of Ii knockout (KO) mice largely depends on the H-2 haplotype of the mouse strain in study (23) and also the cell types investigated, (24, 25). In mice bearing the MHC class II of H-2b haplotype, Ii deficiency results in a dramatically reduced surface expression of MHC class II molecules due mainly to the decreased rates of post-ER transport (14, 15, 21). Even the small numbers of mature class II molecules that reached the cell surface had different biochemical properties. They exhibited reduced migration in the SDS-polyacrylamide gels, were capable of binding to a vast array of peptides, and were conformationally different from mature class II molecules expressed in wild-type mice (14). These observations indicated that class II molecules assembled and expressed in the absence of Ii are floppy due to the absence of any peptide bound to the peptide-binding groove or due to the presence of peptides that could be displaced very easily.
As a sequel to reduced surface expression of MHC class II molecules, absence of Ii results in severe defects in positive selection and maturation of CD4+ T cells (14, 20, 21, 26, 27). Processing of exogenous Ags and subsequent presentation to CD4+ T cells is also defective in this strain of mice (20). However, the class II molecules in Ii-/- mice with H-2k or H-2d haplotypes did not show any evidence for the expression of floppy class II molecules (23). These mice strains expressed comparable levels of surface class II molecules which were conformationally similar to mature class II molecules from wild-type mice with functional Ii and were capable of binding to peptides very efficiently. As a result, BALB/c mice lacking Ii had relatively normal positive selection of CD4+ T cells and responded efficiently to protein Ags (28). B cell maturation has also been reported to be inefficient in mice lacking Ii (28, 29, 30).
Although several studies have addressed the accessory role of the Ii in the proper assembly and expression of murine class II molecules, the extent of Ii participation in the expression of HLA class II (especially DQ8), is not fully known. Understanding the expression and function of DQ8 is crucial not only because it is strongly associated with certain autoimmune diseases such as type 1 diabetes and celiac disease (31), this would also shed some light on the interaction between Ii and DQ8 from the in vivo standpoint. As allotypic variations exist within the murine class II molecules on Ii dependence, the role of Ii in assembly, expression, and function of a human class II molecule was determined using HLA-DQ8 and -DQ6-transgenic mice with disrupted Ii gene.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
BALB/c and C57BL/10 (B10) mice originally came from The Jackson
Laboratory (Bar Harbor, ME). Generation of HLA-DQ8-transgenic mice
lacking endogenous class II molecule (Ao) has
been described earlier (32, 33).
HLA-DQ8.Ii-/- mice were generated as follows.
The Ii KO mice on B10.D2 background (produced by Viville et al., Ref.
15 ; and obtained through J. Miller, University of Chicago,
Chicago, IL) were crossed with HLA-DQ8-transgenic mice. The
heterozygous offspring were intercrossed and the
Ii-/- littermates expressing DQ8 were selected
and intercrossed for several generations to obtain
DQ8.Ii-/- mice. HLA-DQ8 mice with and without
Ii are hereafter referred to as DQ8.Ii+/+ and
DQ8.Ii-/-, respectively. Lines lacking the
HLA-DQ8 transgene and Ii were also maintained in the mouse colony.
Similarly, mice transgenically expressing HLA-DQ6 (32)
were bred to Ii-/- mice to generate
Ao.DQ6.Ii-/- mice. The
HLA-transgenic mice do not express any endogenous mouse class II Ags
due to introduction of the inactivated H2-A gene by
mating to the Ao mice (32). Mice
bred in the barrier facility were moved to conventional facility
following weaning and used at 8–12 wk of age unless specifically
mentioned.
Abs and reagents
The following Abs were used for FACS analysis. Anti-TCR V8.1,
2, 3 (F23.1), anti-HLA-DQ (IVD12, HB-14), and L227 (HB-96) were
obtained from American Type Culture Collection (ATCC, Manassas, VA).
Anti-TCR V8 (F23.1), anti-CD4 (RM 4.5), anti-CD8 (53.67),
anti-CD74 (In-1), anti-CD23 (B3B4), anti-mouse IgM
(R6-60.2) and anti-mouse IgD (11-26c.2a) were purchased from BD
PharMingen (San Diego, CA). FITC-conjugated secondary Abs were obtained
from Accurate Chemicals and Scientific (Westbury, NY).
Intracellular staining for detection of Ii
Absence of the Ii was determined by intracellular staining for Ii using FITC-labeled Ab as per the following procedure. PBMC obtained by centrifugation over Ficoll-Paque (Pharmacia, Uppsala, Sweden) were washed with PBS containing 1% BSA and fixed with 4% paraformaldehyde on ice for 20 min. Subsequently, the cell pellet was resuspended in permeabilization buffer (0.1% saponin (Sigma-Aldrich, St. Louis, MO) in PBS containing 1% heat-inactivated FCS and 0.1% sodium azide) for 10 min on ice. Cells were incubated with FITC-labeled Ab for 30 min on ice, washed, and analyzed cytofluorometrically.
Superantigen (SAg) injections and response
Staphylococcal enterotoxin B (SEB; Sigma-Aldrich) was dissolved
in endotoxin-free PBS at a concentration of 1 mg/ml (This preparation
contains only 
25% of protein. The concentration of SEB indicated in
all the experiments represents the initial concentration and not the
final corrected concentration.) For analyzing in vivo T cell expansion,
mice received a single i.p. injection of 50 µg of SEB in 200 µl of
PBS. Single-cell suspensions of splenocytes from naive or SEB-primed
mice were depleted of RBC by buffered ammonium chloride lysis. Cells
were cultured in HEPES-buffered RPMI 1640 containing 5%
heat-inactivated horse serum, serum supplement, and antibiotics
streptomycin and penicillin at the rate of 10 x
105 cells/well in 100-µl volumes in 96-well
flat-bottom tissue culture plates. Cells were cultured with equal
volumes of either Con A (Sigma-Aldrich) or SEB (2.5 and 1 µg/ml,
respectively, unless specifically indicated) for a total of 48 h.
Tritiated thymidine (NEN/PerkinElmer Life Sciences, Boston, MA) was
added to the wells (1 µCi/well) during the last 18 h of culture.
At the end of the assay, the cells were harvested using a plate
harvester (Brandel, Gaithersburg, MD) and the DNA was captured on
filter mats. Incorporated radioactivity was determined using an
automated counter (Microbeta; PerkinElmer Wallac, Gaithersburg,
MD).
In some assays, irradiated splenocytes were used as APCs. For this,
single-cell suspensions of splenocytes were prepared as above and were
irradiated at 3000 rad. The responder cells consisted of
CD3+ T cells enriched from spleens of B10 mice or
TCR-transgenic mice expressing TCR V8 either on
CD4+ (DO11) or CD8+
(lymphocytic choriomeningitis virus (LCMV)-specific) T cells
(courtesy of Dr. L. Pease, Rochester, MN) using mouse T cell
enrichment columns (R&D Systems, Minneapolis, MN) as per the
manufacturer’s protocol. Enriched CD3+ T cells
obtained as above were cultured (5 x 105
cells/well) with equal numbers of irradiated APCs from either
DQ8.Ii-/- or DQ8.Ii+/+
mice in the presence or absence of varying concentrations of SEB for a
total of 72 h, and cell proliferation was determined by thymidine
incorporation.
Peptide immunization and lymphoproliferation assay
A previously described peptide (P44) encompassing aa 554–573 from human type II collagen, which is highly immunogenic in DQ8 mice (33), was emulsified in CFA and injected s.c. at the base of the tail and footpads such that each mouse received 100 µg of the peptide. Seven days later, mononuclear cell suspensions were prepared from draining lymph nodes and cultured in either medium alone or different concentrations of P44 in 96-well flat-bottom tissue culture plates in triplicates (1 x 106/well) for a total of 48 h. Cell proliferation was determined by thymidine incorporation as described above.
Ag presentation assays
HLA-DQ8-transgenic mice were immunized with P44 and lymph node mononuclear cells were isolated 7 days later as described above. CD3+ T cells from pooled lymph node mononuclear cells were enriched by depleting B cell and macrophages using columns as described earlier. The enriched CD3+ T cells (5 x 105 cells/well) were cultured with equal numbers of irradiated splenocytes from either DQ8.Ii+/+ or DQ8.Ii-/- mice as APCs for a total of 72 h in the presence of Con A or various concentrations of P44. Cell proliferation was determined by thymidine incorporation as described above.
In some assays, irradiated splenocytes from DQ8.Ii+/+ and DQ8.Ii-/- (5 x 106 cells/ml) mice were used to present P44 to a T cell hybridoma (1 x 106 cells/ml) specific for this peptide. Cells were cultured in triplicate wells in 100-µl volumes. Culture supernatants were collected at 18 and 36 h later and IL-2 present in the supernatant was quantified by ELISA as described below. The following controls were also included. T cell hybridoma cultured with irradiated APCs without any peptide, T cell hybridoma, and irradiated APCs cultured with peptide alone without APCs and hybridoma, respectively.
Cytokine analysis
Splenocytes from naive mice were cultured in 24-well plates for 72 h with SEB (1 µg/ml) or left unstimulated. Cytokines present in the culture supernatants were quantified by sandwich ELISA as described earlier (34). Cytokines present in the sera were also determined by routine sandwich ELISA as per published protocol (34).
Induction of toxic shock with SEB
Mice received two i.p. injections of SEB (100 µg in 200 µl of PBS) 48 h apart and were closely monitored for the symptoms of shock. In the D-galactosamine (D-gal)-sensitized toxic shock model, mice were injected i.p. with 20 mg of D-gal (Sigma-Aldrich) followed by SEB (20 µg/mouse), each in 200 µl of PBS.
Estimation of serum IgG levels
Mice were bled at different time points and the concentration of serum IgG was estimated by standard ELISA. Briefly, ELISA plates were coated with goat anti-mouse IgG (10 µg/ml) in sodium bicarbonate buffer as the capture Ab overnight at 4°C. Following blocking with 3% BSA for 2 h, adequately diluted serum samples were added to the wells in duplicates and incubated further for 2 h. Following a washing step, peroxidase-conjugated goat anti-mouse IgG was added as the detection Ab. Mouse IgG captured on the plates was revealed by adding the substrate tetramethylbenzidine. Concentration of IgG present in the sera was extrapolated from the standard curve generated using the purified mouse IgG (Sigma-Aldrich).
Statistics
The statistical significance of the results was determined by using GraphPad Prism software (version, 3.0a; GraphPad, San Diego, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Analysis of the splenocytes from DQ8.Ii-/-
mice by flow cytometry using the Ab IVD12 revealed reduced surface
expression of HLA-DQ8 when compared with the wild-type DQ8-transgenic
mice (Fig. 1
, a–c). Screening
with other HLA class II-specific Abs such as IVA12 (HB145, anti-DR,
-DP, -DQ; ATCC) and 93.F10 (HB-180, anti-DR, -DQ; ATCC) by flow
cytometry also showed similar results (data not shown). Therefore, for
further analysis only IVD12 was used. Even the percentage of cells
expressing DQ8 was reduced in Ii-/- mice (Fig. 1d). Expression of DQ8 remained at low levels when mice of
different age groups were tested, indicating that expression of DQ8 in
Ii-deficient mice does not change with age (data not shown). Similarly,
we could not find any difference in the expression profile of DQ8
between the mice raised in the conventional animal facility vs the ones
reared in the barrier facility (data not shown). This again implies a
minimal influence of the environment on DQ8 expression in
Ii-/- mice. Moreover, culturing splenocytes
from DQ8.Ii-/- mice with LPS did not result in
increased surface expression of class II molecules (data not
shown).
|
As mice lacking functional Ii have reduced surface expression of
class II molecules, we studied the development of
CD4+ T cells in these mice. As predicted,
DQ8-transgenic mice lacking Ii had significantly reduced numbers of
mature CD4+ T cells in the periphery with a
slight increase in CD8+ T cells (Fig. 2a). Defective
CD4+ T cell maturation in
Ii-/- mice was persistent and did not improve
with age (Fig. 2, a and b, and additional data
not shown). Similarly, raising the animals in the conventional facility
did not rescue CD4+ T cell maturation (Fig. 2c).
|
, d and
e). In vitro response to exogenous SAg SEB
SEB-induced proliferation of splenocytes from
DQ8.Ii+/+-transgenic mice was severalfold higher
than that of DQ8.Ii-/- mice at all doses of SEB
tested (Fig. 3a). We also
compared the proliferative responses of splenocytes from
DQ8.Ii-/- mice with that of B10 mice to both
SEB and to a nonspecific mitogen, Con A. As shown in Fig. 3b, the proliferative response of
DQ8.Ii-/- splenocytes to SEB was significantly
higher than that of B10 mice and this difference was more pronounced at
lower dose of SEB. This response was DQ8 dependent as the proliferative
response was almost absent in Ii-/- mice
lacking the DQ8 transgene (Fig. 3b). Defective proliferation
of B10 splenocytes was specific for SEB as they proliferated very
efficiently to Con A. This is because HLA class II can present SEB more
efficiently than H2-Ab (35).
|
(as a marker of the Th1 response) were measured
in addition to IL-2. As depicted in Fig. 3c,
DQ8.Ii-/- mice produced significantly less
amounts of all the cytokines tested, but the extent of this defect
varied with the cytokine. IL-2 and IL-4 levels were more affected than
IFN- or IL-10 (Fig. 3c).
In some experiments, irradiated splenocytes from
DQ8.Ii-/- mice were used to present SEB to
enriched CD3+ splenocytes from either B10 mice or
from mice transgenically expressing TCR V8 (Fig. 4). When CD3+
splenocytes from B10 mice were used as responders, defective SAg
presentation by DQ8.Ii-/- was more apparent at
lower concentrations of SEB (Fig. 4a). The same was true
when the enriched CD3+ T cells from mice
expressing TCR V8 on CD4+ T cells were used
(Fig. 4b). However, when the enriched
CD3+ T cells from mice expressing TCR V8 on
CD8+ T cells were used as responders, very little
proliferation could be observed with DQ8.Ii-/-
APCs (Fig. 4c). The results indicate that 1)
DQ8.Ii-/- mice are compromised in their ability
to respond to the bacterial SAg SEB both in terms of cell proliferation
and cytokine production when compared with their Ii-sufficient
counterpart and 2) DQ8.Ii-/- mice still express
residual levels of DQ8 and are thus capable of inducing proliferation
of responder cells in response to SEB (which is superior to murine
class II, H2-Ab).
|
To study in vivo T cell activation in SEB-primed Ii-deficient and
-sufficient DQ8-transgenic mice, splenocytes were collected at
different time points and analyzed by flow cytometry. The percentage of
total CD4+ and CD8+ T cells
and the percentage of T cells expressing TCR V8 within these
populations was determined (Fig. 5). SEB
caused massive activation resulting in expansion of both
CD4+ and CD8+ T cells
bearing TCR V8 family in
HLA.DQ8.Ii+/+-transgenic mice. There was a 5-fold
increase in both CD4+ and
CD8+ T cells bearing TCR V8 by day 3 after SEB
injection. The number of TCR V8-bearing CD4+
cells remained elevated (4-fold) even by day 7, while the TCR
V8-bearing CD8+ T cell numbers were only
slightly above the pretreatment values at day 7. As
DQ8.Ii-/- mice have very low
CD4+ T cell numbers, there was not a significant
expansion of CD4+ T cells in these mice at any
time point. However, there was a definite and comparable increase in
TCR V8-bearing CD8+ cells in these mice at day
3 followed by a deletion in TCR V8-bearing
CD8+ cells at day 7 similar to that seen in
DQ8.Ii+/+ mice. The results indicate that even
though DQ8.Ii-/- mice express very low levels
of surface DQ8, leading to improper positive selection of
CD4+ T cells, they express adequate levels of DQ8
to present the SAg SEB and to drive the expansion of SEB-reactive
CD8+ T cells. Proliferation of splenocytes from
DQ8.Ii-/- mice in response to SEB and the
ability of APCs from DQ8.Ii-/- mice to activate
T cells bearing TCR V8 support this observation.
|
We have previously shown that HLA-DQ8-transgenic mice are
susceptible to SEB-induced toxic shock even without sensitization with
D-gal (34). Because
DQ8.Ii-/- mice expressed very low levels of
surface DQ8 and had fewer CD4+ T cells in the
periphery, we wondered whether the absence of Ii rendered
HLA-DQ8-transgenic mice resistant to SEB-induced shock. Although all of
the DQ8.Ii+/+ mice succumbed following the second
SEB injection, none of the DQ8.Ii-/- mice died
(Table I). We then studied SEB-induced
mortality in D-gal-sensitized mice. As shown in Table I,
50% of the DQ8.Ii-/- mice were still
susceptible to SEB-induced toxic shock while none of the
Ii-/- mice lacking the HLA class II transgene
(Abo.DQ8-.Ii-/-) and none of the B10 mice
succumbed, indicating clearly that DQ8 is important for this
process.
|
and IFN- by ELISA. We could not detect any TNF-
even in sera of DQ8.Ii+/+ mice that succumbed to
SEB-induced shock (data not shown), whereas we could detect very high
levels of IFN- in these sera at 3 h following SEB injection
(Fig. 6).
DQ8.Ii-/- mice had significantly less IFN-
in the sera than Ii-sufficient mice, implying a possible pathogenic
role for IFN- in SEB-induced lethal shock.
|
We also tested the ability of DQ8.Ii-/-
mice to mount a T cell response to P44, which has been shown by us
previously to be immunogenic in DQ8 mice (33). Although
lymph node cells from peptide-immunized DQ8.Ii+/+
could proliferate efficiently upon restimulation with the same peptide,
no response could be elicited in cells obtained from
DQ8.Ii-/- mice immunized with the same peptide,
indicating severe defects in mounting a CD4+ T
cell-dependent immune response (Fig. 7a). However, irradiated
splenocytes from naive DQ8.Ii-/- mice were
capable of presenting peptides to T cells obtained from DQ8 mice
immunized with P44 (Fig. 7b). We also studied the ability of
splenocytes from Ii-sufficient and -deficient mice to present peptide
to a peptide-specific T cell hybridoma. Splenocytes from
Abo.DQ8.Ii-/- mice were less efficient in
eliciting IL-2 production by the T cell hybridoma at 18 h.
However, by 36 h, Abo.DQ8.Ii-/-
splenocytes could induce IL-2 production though not to the same extent
as Abo.DQ8.Ii+/+ splenocytes (Fig. 7c). The results indicate that even though Ii deficiency
results in reduced expression of DQ8 (which leads to poor positive
selection of CD4+ T cells and compromises the
ability to mount a CD4+-mediated T cell
response), the Ag-presenting capacity of the APCs is still maintained
albeit in reduced efficiency.
|
We also studied the role of Ii on expression/function of another
related HLA class II molecule, DQ6. Expression of DQ6 was analyzed
using two different Abs, IVD12 and L227. (We have observed that L227
can recognize the -chain encoded by DQ6*0601 in addition to its
reactivity to the DR -chain and can thus be used to monitor surface
expression of DQ6.) Irrespective of the Ab used,
DQ6.Ii-/- mice always showed poor expression of
DQ6 when compared with DQ6.Ii+/+ mice (Fig. 8a), thus proving that even
DQ6 is dependent on Ii for its expression. As a reflection of poor DQ6
expression, CD4+ T cell development was also
compromised in DQ6.Ii-/- mice (Fig. 8b). Poor proliferative response to SEB underscores these
defects (Fig. 8c). Overall, it appears that Ii is necessary
for proper expression of HLA-DQ6 and -DQ8 molecules.
|
Because the Ii has been shown to play an important role in
maturation of B cells, we estimated the percentage of mature B cells in
peripheral lymphoid organs of DQ8.Ii-deficient mice.
DQ8.Ii-/- mice had significantly reduced
numbers of CD23high (Fig. 9a) and
IgDhigh cells (Fig. 9b) in the spleen
and the difference was more pronounced in older mice. Similar results
were obtained when lymph nodes from these mice were analyzed. However,
mature B cells were not completely absent in Ii-deficient mice.
Although DQ8.Ii-/- mice had comparable numbers
of B220+ cells at 10 wk of age, indicating that
there was no defect as such in the development of B cell precursors, at
40 wk of age there was a 2-fold reduction in
B220+ cells, implying that B cell development
could be compromised in older DQ8.Ii-/- mice
(Fig. 9c). (B220 is a pan-B cell marker expressed from pro-B
cell to mature B cell stages.) Furthermore, as an indicator of mature B
cell function, the serum IgG level in naive
DQ8.Ii-/- mice was quantitated at two time
points. DQ8.Ii-/- mice consistently had
significantly less serum IgG levels when compared with the wild-type
DQ8-transgenic mice (Fig. 9d). Immunizing
DQ8.Ii-/- mice with a known Ag in CFA also did
not result in an increase in CD23+,
IgM+, or IgD+ cells (Fig. 10, a–d) nor an increase in
CD4+ T cells in the draining lymph nodes (Fig. 10, e and f).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The phenotype of HLA-DQ8.Ii-/- mice was similar
to that of H-2b mice lacking functional Ii in
that surface expression of fully mature HLA-DQ8 was significantly
reduced. Surface expression of DQ8 was determined by flow cytometry
using the Ab IVD12 (44). There exists a possibility that
in the absence of Ii, the DQ8 heterodimer might exist in
different conformation and hence IVD12 is not able to detect them, but
the normal level of DQ8 is still expressed in
DQ8.Ii-/- mice. However, this is unlikely as
the expression profile remained similar even when other Abs were used
for screening. Moreover, defective expression of DQ6 as assessed by two
different Abs underscores the importance of Ii on expression of HLA-DQ
molecules (DQ6 and DQ8). CD4+ T cell numbers are
also significantly reduced in these mice. Positive selection of
CD4+ T cells in the thymus depends on the
presence of functional class II molecules. Significantly reduced
numbers of CD4+ T cells both in the thymus of
DQ8.Ii-/- (not analyzed for
DQ6.Ii-/-) and in the periphery of both
DQ8.Ii-/- and DQ6.Ii-/-
mice clearly indicate a significant reduction in the functional class
II molecules in Ii-deficient mice. It is of interest to note that
HLA-DM molecules are capable of chaperoning DR molecules from ER
through Golgi even in the absence of Ii in some cell lines (45, 46). However, this does not seem to be an efficient pathway in
vivo in mice. This could be due to differences between mouse H-2M
molecules and human HLA-DM or it could be due to inherent differences
in the class II molecules in question, DR vs DQ8.
We further studied the functionality of DQ8/DQ6 in Ii-deficient mice by
testing its ability to present the bacterial SAg SEB. SAg, a family of
microbial proteins, are strong polyclonal activators of T and B
lymphocytes. The T cell SAg differ from mitogens in that activation by
SAg is MHC class II dependent and TCR mediated. They differ from
conventional Ags in that T cell activation by SAg is MHC dependent,
peptide nonspecific, and CD4/CD8 coreceptor independent. Although a
processed nominal Ag is presented either by MHC class I or class II
molecules to CD8+ and CD4+
T cells, respectively, SAg in their native form bind to MHC class II
molecules outside the peptide-binding groove and vigorously activate
both T cell subsets bearing certain TCR V families (reviewed in Ref.
47). Because class II molecules are mandatory for
eliciting a response to SAg, the poor proliferative response of
splenocytes from DQ6/DQ8.Ii-/- mice further
confirms that there are defects in expression of the fully functional
class II molecules. It is of interest to note that in spite of the
reduced expression of class II molecules in
DQ8.Ii-/- mice, the response to SEB was
significantly higher than that of wild-type B10 mice expressing normal
levels of class II and CD4+ T cells. This is due
to the fact that HLA class II molecules present SEB more efficiently
than their murine counterpart (35).
Defective T cell response in DQ6/DQ8Ii-/- mice
to SEB could be attributed to either low CD4+ T
cell numbers (the responders) or to reduced expression of surface class
II molecules (the presenters) and most likely the combination of both
the defects. Less efficient presentation of SEB by
DQ8.Ii-/- APCs to either T cells from B10 mice
or from TCR-transgenic mice expressing V8 on
CD4+ T cells reveal defective functioning of MHC
class II molecules (DQ8) in the absence of Ii.
DQ8.Ii-/- APCs were also less efficient in
activating T cells from TCR-transgenic mice expressing V8 on
CD8+ T cells. Comparatively, the defects
in presentation of SEB by DQ8.Ii-/- APCs were
more pronounced when CD8+ T cells were used as
responders. This could be due to possible differences in the responder
cell types (CD4+ vs CD8+ T
cells). Because the nature of the T cell response to SAg depends on
several factors such as TCR V usage (47), TCR V
usage (48), ligand density (49), the peptide
repertoire bound by class II molecules, etc. (50, 51), it
is beyond the scope of the present investigation to explore these
differences. The main conclusion from this set of experiments is that
in the absence of Ii, the efficiency of presentation of SAg by DQ8 is
reduced but not completely lost.
Even though DQ8 molecules in Ii-/- mice were
less efficient in inducing proliferation of CD8+
T cells expressing the LCMV-specific TCR V8 transgene, in vivo
administration of SEB resulted in an increase in TCR V8-expressing
CD8+ T cells in DQ8.Ii-/-
mice which was comparable to that seen in
DQ8.Ii+/+ mice. However, the
CD4+ T cell population did not show any
significant change in the DQ8.Ii-/- mice. This
clearly indicates that even the reduced levels of surface DQ8 is still
sufficient to drive proliferation of TCR
V8+CD8+ T cells in
Ii-/- mice. Activation of
CD8+ T cells by the residual DQ8 might explain
the proliferation and cytokine production still seen in
DQ8.Ii-/- cultures.
SAg-induced toxic shock is the major pathology seen mainly in humans
resulting from massive T cell activation and concomitant cytokine
production (52). Mice are generally considered to be
resistant to SEB-induced shock because injection of SEB alone is rarely
lethal in mice, except at very high doses and only in certain mouse
strains and not in others (53). We have shown previously
(34) and also in the present study that DQ8-transgenic
mice are highly susceptible to SEB-induced toxic shock even without
D-gal sensitization, and the pathology is probably due to
high production of IFN-. However, DQ8.Ii-/-
mice were completely resistant to this procedure. We could consistently
detect very high levels of IFN- in the sera of
DQ8.Ii+/+ mice but not in
DQ8.Ii-/- mice following injection with SEB.
Protection of DQ8.Ii-/- mice from SEB-induced
shock is probably due to low levels of surface DQ8 and low numbers of
CD4+ T cells which are the major producers of
IFN- following activation with SEB. However, pretreatment with
D-gal, which sensitizes the hepatocytes to apoptosis
induced by TNF- (53), rendered 50% of
DQ8.Ii-/- mice still susceptible to SEB-induced
shock while B10 mice and Ii-/- mice lacking DQ8
were completely resistant (Table I). Mortality in
D-gal-sensitized DQ8.Ii-/- mice
could be due to cytokines produced by CD8+ T
cells, which as discussed earlier, are activated by the low levels of
DQ8 still expressed in these mice.
DQ8.Ii-/- mice closely resembled Ii KO mice
expressing endogenous H-2b class II molecules
even with respect to mounting an immune response to peptide Ags
(21). Priming with an immunogenic peptide elicited poor
secondary T cell responses in DQ8.Ii-/- mice.
Interestingly, the comparable efficiency with which the irradiated
DQ8.Ii-/- APCs presented the peptide to primed
T cells from wild-type DQ8 mice indicates that even in the absence of
Ii, low levels of DQ8 reach the cell surface and are still capable of
presenting exogenously added peptides (4, 21, 45, 54). The
ability of Abo.DQ8.Ii-/- APCs to activate the T
cell hybridoma by presenting the specific peptide further supports this
hypothesis. The time delay in activating the T cell hybridoma
effectively by DQ8.Ii-/- APCs is probably due
to low levels of surface DQ8 molecules in
Abo.DQ8.Ii-/- mice and hence it might take a
longer time to achieve the ligand density/threshold to activate the
hybridoma. As discussed earlier, SEB-induced proliferation of
DQ8.Ii-/- splenocytes in vitro and the
comparable expansion of CD8+ T cells expressing
TCR V8 following in vivo administration of SEB in
DQ8.Ii-/- mice lend support to this
hypothesis.
Another interesting concept that emerged from previous studies of
Ii-deficient mice is its role in maturation of B lymphocytes.
Ii-deficient mice have defective differentiation of B cells from the
immature stage to the mature stage (28, 29). This results
in elevated numbers of immature B cells which are characterized by low
levels of surface IgD and CD23. Recently, Matza et al.
(30) have elucidated the role of the Ii in B cell
maturation. They have shown that the Ii induces B cell maturation by
up-regulating transcription mediated by the NF-
B p65/RelA pathway.
However, the entire molecular event involved in this process is still
undefined. Even in the present investigation, we observed that naive
DQ8.Ii-/- showed similar defects in B cell
maturation when compared with DQ8.Ii+/+ mice and
the serum IgG levels were significantly low but not absent in these
mice. Similar results have been reported earlier. Infection of
Ii-/- mice with viruses elicited significant
but reduced IgG response when compared with that of wild-type mice
(41). Even in the present study, immunization with an
immunogenic peptide did not reverse this defect.
A recent study has shown that nonobese diabetic mice express very high levels of I-Ag7 associated with CLIP, implying a strong interaction of Ii with I-Ag7 and that Ii might be crucial for efficient expression of I-Ag7 (55). The structural similarity between I-Ag7 and HLA-DQ8 (56) suggests that the latter might also strongly associate with Ii. Elution of naturally present Ii-derived peptides from DQ8-expressing cells supports this hypothesis (57). Therefore, Ii deficiency might have resulted in severe defects in DQ8 expression. At this juncture, it should be noted that transgenic mice expressing HLA class II as the only restricting molecule do mount a normal immune response and are indeed susceptible to several autoimmune diseases indicating the normal functioning of human class II in the murine system (58). This rules out the possible effects of heterologous interaction between human class II and murine accessory molecules, Ii and H-2M. Thus, in conclusion, our study for the first time shows that expression of fully functional HLA-DQ6 and -DQ8 in mice is highly dependent on the Ii chain and possibly Ii could play a similar role in humans. In addition, our study reconfirms the partial defects in B cell maturation in Ii-deficient mice.
|
Acknowledgments |
|---|
o mice. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Chella S. David, Department of Immunology, Mayo Clinic, 200 First Street, SW, Rochester, MN 55905. E-mail address: davic4{at}mayo.edu
3 Abbreviations used in this paper: Ii, invariant chain; SAg, superantigen; SEB, staphylococcal enterotoxin B; CLIP, class II-associated invariant peptide; LCMV, lymphocytic choriomeningitis virus; D-gal, D-galactosamine; ER, endoplasmic reticulum; KO, knockout; SP, single positive.
Received for publication January 22, 2002. Accepted for publication June 12, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B-dependent transcription program. J. Biol. Chem. 276:27203.-chain in superantigen recognition. Immunol. Res. 15:98.[Medline]
This article has been cited by other articles:
|
|
 |
|
  E. Cook, X. Wang, N. Robiou, and B. C. Fries Measurement of Staphylococcal Enterotoxin B in Serum and Culture Supernatant with a Capture Enzyme-Linked Immunosorbent Assay Clin. Vaccine Immunol., September 1, 2007; 14(9): 1094 - 1101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Rajagopalan, R. Patel, S. V. Kaveri, and C. S. David Comment on: Human intravenous immunoglobulin for experimental streptococcal toxic shock: bacterial clearance and modulation of inflammation J. Antimicrob. Chemother., January 1, 2007; 59(1): 157 - 159. [Full Text] [PDF]
|
|
|
|
 |
|
 G. Rajagopalan, K. Iijima, M. Singh, H. Kita, R. Patel, and C. S. David Intranasal Exposure to Bacterial Superantigens Induces Airway Inflammation in HLA Class II Transgenic Mice Infect. Immun., February 1, 2006; 74(2): 1284 - 1296. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Rajagopalan, M. M. Sen, and C. S. David In Vitro and In Vivo Evaluation of Staphylococcal Superantigen Peptide Antagonists Infect. Immun., November 1, 2004; 72(11): 6733 - 6737. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. Dissanayake, J. A. Thompson, J. J. Bosch, V. K. Clements, P. W. Chen, B. R. Ksander, and S. Ostrand-Rosenberg Activation of Tumor-specific CD4+ T Lymphocytes by Major Histocompatibility Complex Class II Tumor Cell Vaccines: A Novel Cell-based Immunotherapy Cancer Res., March 1, 2004; 64(5): 1867 - 1874. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Rajagopalan, Y. C. Kudva, L. Chen, L. Wen, and C. S. David Autoimmune diabetes in HLA-DR3/DQ8 transgenic mice expressing the co-stimulatory molecule B7-1 in the {beta} cells of islets of Langerhans Int. Immunol., September 1, 2003; 15(9): 1035 - 1044. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
