| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-Subunit Promotes T Cell Tolerance to the Immunodominant Gastritogenic Determinant1
* Department of Biochemistry and Molecular Biology, University of Melbourne, and
Department of Pathology and Immunology, Monash University Medical School, Melbourne, Victoria, Australia
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit (H/K) is required for the onset of experimental
autoimmune gastritis in BALB/c mice. The extent to which endogenous
H/K contributes toward the tolerance of the H/K-specific T cell
repertoire in normal individuals is not known. By comparison of T cell
responses in H/K-deficient (o/o) and H/K-expressing BALB/c
mice, in this work we show that the endogenous H/K autoantigen plays
a major role in the tolerance of pathogenic H/K-specific T cells.
First, T cell-dependent Ab responses to the H/K Ag were enhanced in
H/K ATPase-immunized H/K-deficient mice compared with wild-type
mice. Second, peptide immunization experiments indicated that immune
responses to the major gastritogenic epitope of the H/K ATPase, namely
H/K253–277, were significantly more vigorous in
H/K-deficient mice compared with wild-type mice. Third,
unfractionated splenocytes from H/K-deficient mice, but not
H/K-expressing mice, induced autoimmune gastritis after adoptive
transfer to BALB/c nude mice. The enhanced responses to H/K in
H/K-deficient mice were shown to be intrinsic to
CD4+CD25- T cells rather than a change in
status of CD4+CD25+ regulatory T cells. We
conclude from these studies that the H/K-specific T cells in
wild-type mice represent the residue of a T cell repertoire, directed
toward a single determinant, that has been subjected to partial
tolerance induction. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Experimental autoimmune gastritis represents an excellent model of
human pernicious anemia, an organ-specific autoimmune disease of the
stomach (8). Experimental autoimmune gastritis can be
elicited in genetically susceptible strains of mice by a variety of
manipulations, including neonatal thymectomy (9, 10, 11) and
immunization with purified gastric H/K ATPase (12, 13).
Autoimmune gastritis has the hallmarks of an inflammatory autoimmune
disease, as the early lesion is characterized by an influx of
CD4+ T cells and macrophages into the stomach
mucosa (14). Adoptive transfer and in vivo depletion
studies of T cell subsets have demonstrated that the development of the
gastric lesion is mediated by CD4+ T cells and
not by CD8+ T cells (15).
Furthermore, a T cell response to the gastric H/K ATPase -subunit
(H/K),3 a subunit
of the abundant H/K ATPase membrane protein in parietal cells, is
necessary for the initiation of gastritis. Mice rendered tolerant to
the H/K, by transgenic expression of the H/K in the thymus,
failed to develop gastritis after neonatal thymectomy
(16), immunization with the mouse gastric H/K ATPase
(17), or adult thymectomy combined with cyclophosphamide
treatment (18). In contrast, expression of the H/K ATPase
-subunit (H/K) in the thymus did not prevent the development of
autoimmunity (17).
Anti-H/K-specific T cells are present in the periphery of normal
BALB/c mice (17, 19). The extent to which endogenous
H/K contributes to the tolerance of the H/K-reactive T cell
repertoire in normal individuals is not known. H/K is expressed
abundantly in the gastric mucosa and also at low levels in the kidney
(20, 21); previously we reported that H/K transcripts
were not detected in the thymus by RT-PCR using RNA from whole thymus
(16). One possibility is that endogenous H/K may not
cause clonal tolerance in BALB/c mice and that ignorance and/or
regulation may be the primary tolerance mechanism to the gastric
H/K. In support of this view, although T cells that have the
potential to induce autoimmune gastritis are present in normal mice,
these T cells cause gastritis only upon manipulation of the immune
system by treatments such as adult thymectomy combined with
cyclophosphamide, by transfer of
CD4+CD25- T cells from
normal BALB/c mice into a T cell-deficient recipient, or by activation
of H/K-specific T cells following immunization with H/K ATPase to
induce the inflammatory disease of the gastric mucosa (13, 17, 18, 22).
We have previously mapped the dominant gastritogenic epitope of H/K
in H-2d mice to residues 253–277
(H/K253–277) based on the following: first, T
cells from gastric H/K ATPase-immunized mice responded to only one of
the complete set of H/K peptides, namely
H/K253–277; second, immunization of mice with
the H/K peptides resulted in a T cell response to only
H/K253–277 (19); third, multiple
immunization with H/K peptides demonstrated that
H/K253–277 was capable of inducing a
mononuclear infiltrate, specifically within the gastric mucosa
(19); and fourth, 20% of
H/K253–277-specific TCR transgenic mice
spontaneously develop autoimmune gastritis, confirming the pathogenic
potential of H/K253–277-specific T cells in
vivo (23). However, immunization of
H-2d mice with H/K ATPase or with
H/K253–277 appears to induce
anti-H/K-specific T cells of low affinity (24),
suggesting that the anti-H/K repertoire present in normal
individuals is of low avidity. To determine whether the T cells in
H/K-expressing mice represent the full H/K-reactive repertoire or
whether they may be the residue of a T cell repertoire that has been
subjected to Ag-specific tolerance induction, in this work we have
compared T cell responses in H/K-deficient mice (o/o) and in
H/K-expressing BALB/c mice. H/K-deficient mice have a mutation in
the first exon of the H/K gene after the third codon
(25) and are thus unable to synthesize any potential T
cell epitopes. H/K-deficient mice are healthy and viable
(25). We show here that H/K-deficient mice have
enhanced T cell responses to H/K, compared with H/K-expressing
mice, clearly demonstrating an Ag-dependent mechanism responsible for T
cell tolerance to the immunodominant epitope of the gastric H/K.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Six- to 12-wk-old BALB/cCrSlc mice were obtained from the Monash
University Central Animal Facility (Clayton, Australia) and BALB/c nude
(nu/nu) mice from the University of New South Wales (Sydney,
Australia). H/K-deficient mice have been previously described
(25) and were backcrossed at least six times onto the
BALB/cCrSlc background. All mice were housed under conventional
conditions at the Monash University Medical School Animal Facility. All
work with animals was performed with approval of the Monash University
Animal Ethics Committee and the University of Melbourne Animal Ethics
Committee.
Cell suspensions
Splenocyte and lymph node suspensions were prepared in 10% FCS in PBS (FCS/PBS) by gently teasing whole spleens through wire mesh (pore size, 200 µm). Cells were collected by centrifugation and then resuspended in 0.19 M ammonium chloride to lyse RBCs. After centrifugation, cells were resuspended in FCS/PBS. Cell density and viability were monitored using ethidium bromide and acridine orange and counted using a hemocytometer.
For isolation of CD25- splenocytes, the spleen cell suspensions were incubated with purified anti-CD25 mAb (clone PC61; 1 µg/108 cells) on a rotating wheel at 4°C for 1 h, washed in FCS/PBS, and transferred to tubes containing washed Dynabeads M-450 sheep anti-rat IgG (eight beads per target cell; Dynal Biotech, Oslo, Norway). The cell/bead mixture was incubated on a rotating wheel at 4°C overnight and the CD25- cells were collected according to the manufacturer’s instructions and analyzed using the anti-CD25 mAb clone 7D4 (BD PharMingen, San Diego, CA) by flow cytometry using a FACScan flow cytometer (BD Biosciences, Mountain View, CA).
Adoptive transfer of splenocytes and CD25- T cells
A total of 3 x 107 splenocytes, or 4 x 107 CD25- cells (in 500 µl of FCS/PBS), from 8- to 10-wk-old animals were administered i.p. into 6-wk-old BALB/c nu/nu mice. Six to 8 wk after transfer, sera were collected and the presence of autoantibodies was examined by ELISA as described (26). Histological examination of the stomachs was also performed to assess the presence of mononuclear cell infiltrates. Autoimmune gastritis was graded on a scale of 0–3 according to the following criteria: 0, normal gastric mucosa; 1, mononuclear cell infiltrate, mainly restricted to submucosa, and no hypertrophy or disruption of gastric units; 2, mononuclear cell infiltrate extending into the mucosa and gastric unit hypertrophy and/or disruption of the gastric units; 3, mucosal mononuclear cell infiltrate or no infiltrate, gastric unit hypertrophy and severe disruption of normal gastric units with depletion of parietal cells and zymogenic cells. Histological analyses were performed blind by two individuals with agreement of scoring in the majority of cases. Data were analyzed by the Mann-Whitney U test.
Purification of mouse gastric H/K ATPase
Gastric H/K ATPase was purified from extracts of mouse stomachs as described (13). The purity of H/K ATPase preparations was assessed by SDS-PAGE under reducing conditions. The gastric H/K ATPase was methanol precipitated before use in in vitro T cell proliferation assays to ensure sterility. Briefly, one part protein solution was added to nine parts ice-cold methanol and precipitated at -20°C for at least 24 h. The precipitate was pelleted by centrifugation, dried, and resuspended in sterile PBS. Protein concentration was determined using a microBSA protein assay (Pierce, Rockford, IL).
Peptides
Overlapping peptides spanning the entire sequence of the mouse
gastric H/K were synthesized by Auspep (Parkville, Australia).
Peptides were synthesized by standard protocols. The location of
individual H/K peptides within the -subunit polypeptide has been
described (19). Peptides 1–20 are 25-mers and peptides 21
is a 14-mer. All peptides have an 11-aa overlap. For some assays the
H/K peptides were divided into five pools, with pool 1 containing
peptides 1–4 (spanning residues 1–67), pool 2 containing peptides
5–8 (spanning residues 57–123), pool 3 containing peptides 9–12
(spanning residues 113–179), pool 4 containing peptides 13–16
(spanning residues 169–235), and pool 5 containing peptides 17–21
(spanning residues 225–294). Peptides were dissolved in sterile water
at 10 mg/ml.
Immunization protocol
Six- to 10-wk-old H/K-deficient mice and heterozygous
littermate mice were immunized once s.c. in the tail base with 30 µg
mouse gastric H/K ATPase emulsified in an equal volume of CFA (Life
Technologies, Grand Island, NY). For peptide immunizations,
H/K-deficient and heterozygous littermates were immunized once,
twice, or three times at three weekly intervals s.c. in the tail base
or back with 10 µg of each peptide or 40–50 µg of peptide pools
emulsified in CFA. Mice were killed 7 days after the final immunization
and inguinal lymph nodes or spleen were removed for T cell
preparations. Sera were collected before commencing the immunization
protocol and when the mice were killed.
For depletion of CD25+ T cells before Ag
immunization, BALB/cCrslc mice were thymectomized at 3 wk of age, and
then at 6 wk 0.25 mg purified PC61 mAb was injected i.p. One day later
mice were immunized s.c. in the tail base with 50 µg of H/K
peptide pools or with 50 µg of hen egg lysozyme (HEL) emulsified in
CFA. Mice were killed 7 days after the immunization and inguinal lymph
nodes were removed for T cell preparations.
T cell proliferation assays
T cell response to Ags was assessed by T cell proliferation
assays using either lymph node or nylon wool-enriched splenic T cell
preparations. T cells (5 x 105) from
immunized H/K-deficient and heterozygous littermate control mice
were incubated with 5 x 105 irradiated
syngeneic splenocytes in the presence or absence of mouse gastric H/K
ATPase (50 µg/ml), peptides (1.9–50 µg/ml), or 3 µg/ml Con A
(Sigma-Aldrich, St. Louis, MO) in 200 µl of DMEM supplemented with
10% FCS, 2 mM glutamine, 100 IU/ml penicillin, 100 µg/ml
streptomycin, and 50 µM 2-ME. Cells were cultured in 96-well
round-bottom tissue culture plates at 37°C in 10%
CO2. Following a 48-h incubation, 1 µCi
[3H]thymidine was added to each well and the
plates were cultured for another 16 h. Cells were harvested onto
Printed Filtermats A (Wallac, Turku, Finland) using a Skatron Micro
Cell Harvester (Skatron Instruments, Lier, Norway) then dried for
1 h at 37°C. Aqueous scintillant was added and cell-associated
[3H]- thymidine was determined using a 1205
Betaplate Liquid Scintillation Counter (Pharmacia, Turku, Finland).
Each assay was performed in triplicate. Data were analyzed by
ANOVA.
ELISA
To detect Abs to H/K peptides, ELISA plates were coated with
peptide pools containing each peptide at a concentration of 0.6 µM
(1.8 µg/ml) in 0.5 M carbonate buffer (pH 9.6) for 6 h. ELISA
was conducted as described (16) and bound anti-peptide
Abs from immunized mice were detected using an anti-mouse IgG
biotinylated Ab (Amersham, Sydney, Australia).
Immunofluorescence
Abs to the H/K ATPase were detected by immunofluorescence on
either frozen mouse stomach sections (4 µm) or insect cell lines, as
described (16, 17). Insect Sf9 cells alone or infected
with recombinant baculoviruses coding for either the rat - or
-subunit of the H/K ATPase (kindly supplied by D. Greenwood, Monash
University) were spotted onto slides and stored desiccated at -70°C
until use. Monolayers were treated for 10 min in 100% methanol at
-20°C and washed for 5 min in PBS at room temperature. Cells and
stomach sections were incubated with 5% FCS/PBS/0.02% sodium azide
for 20 min and washed with PBS/0.05% Tween 20, then incubated for 30
min at room temperature with mouse serum diluted 1/50 (5%
FCS/PBS/0.02% sodium azide) or mouse mAbs 1H9 or 2B6. Monolayers and
sections were washed six times in PBS/0.05% Tween 20 and incubated
with PE-conjugated anti-mouse IgG (diluted 1/200 with 5%
FCS/PBS/0.02% sodium azide; BD PharMingen) and then washed again six
times with PBS/0.05% Tween 20. Sections were mounted in mowiol
and examined on a Bio-Rad 1024 confocal microscopy system (Bio-Rad,
Hercules, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-deficient mice
To determine whether there was a qualitative difference in the
response to the H/K in H/K-deficient mice compared with wild-type
mice, we first examined whether Ab responses could be detected after
immunizing mice with 30 µg mouse gastric membranes enriched for the
H/K ATPase. We have previously shown that normal BALB/c mice will
generate an autoantibody response to the gastric autoantigen only after
multiple immunizations with gastric autoantigen (13). In
this work we immunized H/K-deficient and H/K-expressing
littermate (w/o) mice as a control group once with H/K ATPase-enriched
mouse gastric membranes and analyzed sera 7 days later for the presence
of IgG autoantibodies by immunofluorescence on sections of normal
gastric mucosa (Fig. 1
, A–D). It should be noted that the heterozygous
H/K-expressing littermates have a similar level of H/K protein in
the gastric mucosa as wild-type mice (Ref. 25 and I.
R. van Driel, unpublished observations). All sera (15 of 15)
from immunized H/K-deficient mice stained parietal cells (12 of 15
showed strong parietal cell staining and 3 of 15 showed weak
reactivity), whereas sera from only 1 of 10 H/K-expressing mice
stained (weakly) parietal cells. To determine whether Abs in the sera
were directed to the H/K ATPase subunits, immunofluorescence analysis
was performed using Sf9 cells infected with recombinant baculoviruses
encoding either the H/K or H/K subunits (Fig. 1, E–L). Strong fluorescence staining of infected
Sf9 cells with H/K ATPase subunit-specific mAbs demonstrated that the
insect cells were expressing the H/K ATPase subunits (Fig. 1, G, H, K, and L). Sera from
immunized H/K-deficient mice strongly stained both - and
-subunit-expressing Sf9 cells, whereas only very weak reactivity was
observed with sera from immunized H/K-expressing mice (Fig. 1, E, F, I, and J). No
staining of infected Sf9 cells was observed with sera from unimmunized
H/K-deficient mice (data not shown). These data show that the
H/K-deficient mice respond more vigorously than wild-type mice to
the H/K ATPase, suggesting a difference in the state of tolerance to
this gastric autoantigen.
|
peptides in H/K-deficient mice
To examine the specificity of the T cell responses in
H/K-deficient mice, mice were immunized with overlapping peptides
that spanned the entire sequence of the mouse H/K. All peptides were
25-mers, except the C-terminal peptide, which was a 14-mer, and the
overlap of each peptide was 11 aa. These peptides were divided into
five pools as described in Materials and Methods.
H/K-deficient and H/K-expressing mice were immunized once, s.c.
in the base of the tail, with one of the five peptide pools. Mice were
killed 7 days later, inguinal lymph nodes were removed, and T cell
proliferation assays were conducted. The responses observed in
H/K-deficient and control animals to pools 1–4 were not
significantly different. In contrast, the stimulation indices (SIs) of
lymph node cells from H/K-deficient mice immunized with
peptide pool 5 and challenged with 10 µg/ml of each peptide were
significantly greater than in control animals (H/K-deficient, SI
range = 2.3–6.8 and mean SI = 4; control SI range =
1–2.9 and mean SI = 1.6; p < 0.0001, ANOVA)
(Fig. 2A). When lymph node
cells were challenged with a lower peptide concentration of 2 µg/ml,
the response by T cells from H/K-deficient mice was also significant
(H/K-deficient mean SI = 2.9; control mean SI = 1.4;
p < 0.0053).
|
peptide pool 5 were also assessed in mice
immunized twice with pool 5 peptides. The level of anti-H/K
peptide pool 5 IgG in all the H/K-deficient mice was considerably
higher (titers 4000–8000) than in the normal animals (titers 200–400)
(Fig. 2
B).
To analyze the T cell response to individual H/K peptides, mice were
immunized with peptide pool 5 and T cell proliferation to the
individual peptides of pool 5 (peptides 17–21) was assessed. The T
cell proliferative response to only one peptide,
H/K253–277 peptide (peptide 19), was
significantly greater in the H/K-deficient mice than in the control
animals (Fig. 3). Immunized
H/K-deficient mice had a mean SI of 5.5 for
H/K253–277, whereas H/K-expressing
littermates had a mean SI of 2.3 (p <
0.0005).
|
-deficient mice cause gastritis in nude mice
To determine whether the enhanced immune responses detected above
would result in gastric pathology, splenocytes from H/K-deficient
mice were transferred into BALB/c nu/nu mice. Normally, for
autoimmune gastritis to develop, splenocytes have to be depleted of
regulatory CD4+CD25+ T
cells before transfer to a lymphopenic recipient (27, 28, 29).
As expected, no gastric infiltrate or gastric mucosal disruption was
observed in the five BALB/c nu/nu mice that received
splenocytes from H/K-expressing mice (Fig. 4). However, following transfer of
unfractionated splenocytes from H/K-deficient mice into BALB/c
nu/nu mice, severe autoimmune gastritis resulted in all mice
after 6 wk (Fig. 4). This autoimmune gastritis was characterized by the
presence of autoantibodies to the H/K ATPase, a mononuclear cell
infiltrate in the stomach of the nude mice, and severe disruption of
the gastric units resulting in depletion of parietal and zymogenic
cells and amplification of the number of immature cells, features
typical of this autoimmune disease (30).
|
-deficient mice are intrinsic to
CD4+CD25- T cells
A likely explanation for the elevated responses detected in the
H/K-deficient mice compared with H/K-expressing mice are
differences in the repertoire of anti-H/K T cells. However, an
alternative possibility is that the absence of H/K protein in the
H/K-deficient mice may have an impact on the population of
CD4+CD25+ T regulatory
cells, and the elevated H/K-specific responses observed in
H/K-deficient mice is a consequence of altered T cell regulation. To
determine whether the
CD4+CD25- effector T cell
population from the H/K-deficient mice differs from the effector
population from H/K-expressing mice, splenocyte populations were
depleted of CD25+ cells and the resulting
populations of CD25- T cells (>97.5%) were
transferred to BALB/c nu/nu mice. As expected, autoimmune
gastritis was detected in all mice 6 wk after transfer (Fig. 5). The severity of gastritis was scored
from 0–3, based on criteria defined in Materials and
Methods. Six of seven BALB/c nu/nu mice (86%) that had
received CD25- splenocytes from H/K-deficient
mice showed complete disruption of the gastric mucosa and an extensive
mononuclear infiltrate (grade 3) compared with three of seven mice
(43%) which received CD25- splenocytes from
normal mice (p < 0.05). Thus, the severity of
the gastritis was significantly greater in BALB/c nu/nu mice
that had received CD25- T cells from
H/K-deficient mice compared with those mice that had received
CD25- T cells from H/K-expressing mice. These
results clearly show that there is a qualitative difference in the
population of pathogenic
CD4+CD25- T cells from
H/K-deficient mice compared with normal mice.
|
T cell response in
immunized mice, T cell responses to H/K peptides were determined in
normal mice that were depleted of
CD4+CD25+ T cells. Mice
were thymectomized before treatment with a CD25 depleting mAb to
eliminate repopulation by thymic emigrants. Thymectomized mice were
treated with 0.25 mg of purified anti-CD25 mAb. Analysis of spleen
and lymph node cell suspensions 2 days after anti-CD25 treatment
showed that the CD4+CD25+ T
cell populations were reduced by 75% of the population present in
untreated mice. Furthermore, the
CD4+CD25+ T cell
populations remained at a reduced level (<40%) throughout the course
of the experiment (8 days). Mice were treated with anti-CD25 mAb
and 1 day later were immunized with H/K peptide pool 5. Mice were
killed 7 days later, inguinal lymph nodes were removed, and T cell
proliferation assays were conducted. No difference in T cell responses
to H/K pool 5 was observed between anti-CD25-treated and
untreated mice (SI range 1.2–1.7). Furthermore, strong T cell
responses to the unrelated Ag, HEL, was observed in both HEL-immunized
anti-CD25-treated mice and HEL-immunized untreated mice, indicating
that the anti-CD25 mAb treatment did not inhibit T cell activation.
These results show that
CD4+CD25+ regulatory T
cells in normal mice do not inhibit H/K-specific T cell responses
following immunization in the presence of an adjuvant. Furthermore,
this result is consistent with recent models that propose that the
suppression of CD4+CD25+
regulatory T cells is inhibited under inflammatory conditions
(31), e.g., a response to a pathogen or adjuvant, probably
by engagement of a glucocorticoid-induced TNFR-related gene
family member (32, 33).
Taken together, these data show that the elevated T cell responses in
the H/K-deficient mice compared with H/K-expressing mice are due
to alterations in the H/K-specific
CD4+CD25- effector T cell
population rather than a change in the status of
CD4+CD25+ regulatory T
cells.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
was
essential for the onset of autoimmune gastritis (16), we
were interested to determine whether immune tolerance to this
well-defined and highly abundant gastric autoantigen was established by
an Ag-dependent (deletion or anergy) or independent (ignorance or
regulation) manner. In this work we have demonstrated enhanced T cell
responses to the H/K in H/K-deficient mice compared with
H/K-expressing mice and conclude from these studies that the
endogenous H/K in normal mice plays a critical role in shaping the T
cell repertoire to this gastric autoantigen.
A number of approaches demonstrated enhanced T and B cell responses to
the H/K ATPase in H/K-deficient mice compared with wild-type mice.
First, high-titer IgG anti-H/K ATPase Abs were detected in
H/K-deficient mice after one immunization with mouse gastric H/K
ATPase, whereas in wild-type mice more than two immunizations with
purified gastric H/K ATPase are required to detect an autoantibody
response (13). IgG Abs were detected to both - and
-subunits of the H/K ATPase. It is likely that an
anti-H/K-specific T cell response can drive B cell responses to
both subunits, because we have previously shown that mice tolerant to
the H/K are unable to elicit Ab responses to either subunit
(17). The enhanced Ab responses observed in the
H/K-deficient mice are likely to reflect differences in the
anti-H/K ATPase T cell repertoire rather than the B cell
repertoire, because the gastric H/K ATPase is an intracellular Ag and
therefore is unlikely to promote B cell tolerance in normal
individuals. Second, T cell proliferative responses to the H/K
peptide H/K253–277 in H/K-deficient mice
were 
2.5-fold higher compared with wild-type mice. Thus, endogenous
H/K markedly reduces the T cell repertoire to the
H/K253–277 epitope.
Previously we have shown that the H/K253–277
contains the dominant T cell epitope of the H/K in H/K-expressing
BALB/c mice (19). In this work we demonstrate that T cells
of H/K-deficient mice responded predominantly to the
H/K253–277 peptide (albeit at greater levels
than control animals). Therefore, even in the absence of endogenous
H/K, the T cell response to the H/K is directed primarily toward
a single determinant. The dominance of a single determinant of the
H/K may be due to the existence of only one
H-2d MHC class II binding motif within the H/K
or, alternatively, to cross-reactivity of T cell responses with the
related Na/K -subunit, a protein that shares 30% sequence
identity with H/K and is expressed in all tissues, including the
thymus (21, 34).
The findings here for the gastric H/K ATPase share some similarities
with myelin basic protein (MBP), an autoantigen of the CNS. Analysis of
T cell responses in MBP-deficient mice has also shown a considerable
increase in T cell responses to MBP compared with MBP-expressing mice;
therefore, like H/K, endogenous MBP has a major impact on the
anti-MBP T cell repertoire, possibly by negative selection of the
high-avidity repertoire (35, 36). However, unlike the
situation for the gastric H/K, the immunodominant determinant in
MBP-deficient mice differs from the immunodominant determinant in the
MBP-expressing mice (35). Likewise for myelin proteolipid
protein, endogenous proteolipid protein has been shown to influence
tolerance in a highly epitope-specific manner (37, 38).
The relevance of enhanced responses to
H/K253–277 in H/K-deficient mice to the
development of autoimmunity was demonstrated by the induction of
autoimmune gastritis after adoptive transfer of unfractionated
splenocytes from H/K-deficient mice to BALB/c nu/nu mice.
Potentially pathogenic autoreactive T cells are known to be present in
the periphery of normal mice; however, numerous studies in rodents have
shown that removal of the
CD4+CD25+ T cell population
from the remaining CD4+ T cells is necessary
before they induce disease after adoptive transfer into a lymphopenic
host (10, 22, 39). Consistent with the previous studies,
adoptive transfer of unfractionated splenocytes from normal mice did
not cause autoimmune gastritis. Therefore, the ability of
unfractionated splenocytes from H/K-deficient mice to cause
autoimmune gastritis, with the generation of autoantibodies to the
gastric H/K ATPase, demonstrates that the repertoire of T cells that
induce gastritis has not been subjected to the normal mechanisms of
tolerance in the absence of the endogenous gastric H/K autoantigen.
Because an immune response to the H/K is essential for the onset of
autoimmune gastritis (16), the induction of autoimmune
gastritis in these transfer studies is most likely due to an expanded
or more potent repertoire of anti-H/K T cells from the
H/K-deficient mice.
How does the endogenous H/K protein shape the repertoire of T cell
specific to this gastric autoantigen? Because both wild-type and
H/K-deficient mice respond to a single immunodominant determinant of
the H/K, one likely possibility is that tolerance results from the
removal of high-avidity T cells to the
H/K253–277. Tolerance to H/K may occur in
either the thymus or the periphery. Thymic epithelial cells have been
shown to express a large array of tissue-specific Ags; however,
intrathymic expression of the H/K was not detected by RT-PCR using
whole thymus as a source of RNA (16). Nevertheless, this
does not rule out the possibility that a very small number of
specialized thymic epithelial cells express this gastric autoantigen.
In contrast, peripheral tolerance of high-affinity H/K-specific
cells may be mediated by the presentation of the gastric H/K epitope
by immature dendritic cells in the local draining lymph node of the
stomach as a consequence of normal parietal cell turnover. This
suggestion is consistent with the data that suggest that immature
dendritic cells deliver tolerogenic rather than activation signals
(40, 41). Heath et al. (4) have also
demonstrated that T cells recognizing Ags in peripheral organs can be
deleted subsequent to presentation by bone marrow-derived APCs. Another
possibility is that H/K-deficient mice may lack
CD4+CD25+ regulatory T
cells that specifically control gastric autoimmunity by suppression of
naive anti-H/K T cells. However, two experimental approaches
argue that this is not the case. First, we demonstrated that the
effector CD4+CD25- T cells
from H/K-deficient mice produce more severe autoimmune gastritis
than the effector population from normal mice when transferred into
BALB/c nu/nu mice. Second, depletion of regulatory
CD4+CD25+ T cell
populations from adult BALB/c mice had no affect on H/K T cell
responses in normal individuals. These findings are consistent with the
conclusion that H/K-deficient mice have an enhanced repertoire of
H/K-specific T cells. It remains a formal possibility that the
residual CD4+CD25+
regulatory T cells following depletion had an effect on the
differentiation and expansion of H/K-specific T cells and that
regulatory T cells other than
CD4+CD25+ T cells could be
influencing the H/K-specific T cell responses, although this would
seem unlikely.
These studies are the first to examine the impact of an autoantigen of
a gastric/endocrine organ on the T cell repertoire. We conclude from
these studies that H/K-specific T cells in normal healthy mice
represent the residue of a T cell repertoire, directed toward a single
determinant, that has been subjected to partial tolerance induction.
These findings are likely to have relevance to other autoantigens
targeted in autoimmune diseases of the endocrine/gastric organs.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Paul A. Gleeson, Department of Biochemistry and Molecular Biology, University of Melbourne, Melbourne, Victoria 3010, Australia. E-mail address: pgleeson{at}unimelb.edu.au
3 Abbreviations used in this paper: H/K, H/K ATPase -subunit; H/K, H/K ATPase -subunit; MBP, myelin basic protein; nu/nu, nude; SI, stimulation index; HEL, hen egg lysozyme.
Received for publication January 22, 2002. Accepted for publication June 20, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit in the thymus may explain the dominant role of the -subunit in the pathogenesis of autoimmune gastritis. Autoimmunity 25:167.[Medline]
-subunit. Immunology 96:145.[Medline]
-subunit. Am. J. Physiol. 268:F363. subunit: gene structure and co-ordinate expression with the subunit during ontogeny. J. Biol. Chem. 267:1165.-chains (CD25). J. Immunol. 155:1151.[Abstract]
-subunit. Autoimmunity 33:1.
subunit is required for normal function, development, and membrane structure of mouse parietal cells. Gastroenterology 117:605.[Medline]
subunit gene family: structure of the murine Na,K-ATPase 2 subunit gene. Genomics 11:435.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. E. Snook, P. Li, B. J. Stafford, E. J. Faul, L. Huang, R. C. Birbe, A. Bombonati, S. Schulz, M. J. Schnell, L. C. Eisenlohr, et al. Lineage-Specific T-Cell Responses to Cancer Mucosa Antigen Oppose Systemic Metastases without Mucosal Inflammatory Disease Cancer Res., April 15, 2009; 69(8): 3537 - 3544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. P. Monteiro, J. Farache, A. C. Mercadante, J. A. Mignaco, M. Bonamino, and A. Bonomo Pathogenic Effector T Cell Enrichment Overcomes Regulatory T Cell Control and Generates Autoimmune Gastritis J. Immunol., November 1, 2008; 181(9): 5895 - 5903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kaparakis, K. L. Laurie, O. Wijburg, J. Pedersen, M. Pearse, I. R. van Driel, P. A. Gleeson, and R. A. Strugnell CD4+ CD25+ Regulatory T Cells Modulate the T-Cell and Antibody Responses in Helicobacter-Infected BALB/c Mice. Infect. Immun., June 1, 2006; 74(6): 3519 - 3529. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Allen, S. Read, R. DiPaolo, R. S. McHugh, E. M. Shevach, P. A. Gleeson, and I. R. van Driel Promiscuous Thymic Expression of an Autoantigen Gene Does Not Result in Negative Selection of Pathogenic T Cells J. Immunol., November 1, 2005; 175(9): 5759 - 5764. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Laurie, N. L. La Gruta, N. Koch, I. R. van Driel, and P. A. Gleeson Thymic Expression of a Gastritogenic Epitope Results in Positive Selection of Self-Reactive Pathogenic T Cells J. Immunol., May 15, 2004; 172(10): 5994 - 6002. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and littermate heterozygous (
) mice were
immunized once s.c. in the tail base with 40–50 µg peptide pool
(pool 1, 2, 3, 4, or 5) emulsified in CFA. Seven days later lymph node
cells (5 x 105) from immunized mice were incubated
with 5 x 105 irradiated syngeneic splenocytes in the
presence or absence of the immunizing peptide pool (peptides at 10 or 2
µg/ml) for 48 h, then 1 µCi [3H]thymidine was
added to each well and the cells were cultured for a further 16 h.
Cells were harvested and [3H]thymidine incorporation was
determined. Each data point represents a different mouse. **,
p < 0.0001; *, p <
0.005. B, H/K
, Gastritis is indicated.