| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
,§
Departments of
*
Molecular Microbiology and Immunology,
Pathology, and
Internal Medicine, University of Missouri, Columbia, MO 65212; and
§
Research Service, Department of Veteran Affairs Medical Center, Columbia, MO 65212
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Activation of CD4+ T cells requires two signals, an Ag-specific signal through the TCR and a secondary costimulatory signal between the T cell and APC. TCR ligation in the absence of a costimulatory signal can result in anergy or cell death (8). The major costimulatory molecules expressed by APC are B7.1 and B7.2, which bind to CD28 expressed on T cells (9). A requirement for T cell costimulation for the development of autoimmunity has been observed in several animal models. Treatment with CTL Ag-4 (CTLA-4)-Ig, which blocks both B7.1 and B7.2, suppressed the development of spontaneous autoimmunity in NZB/NZW F1 (10) and nonobese diabetic (NOD) mice (11), as well as experimentally induced autoimmune encephalomyelitis (EAE) (12) and collagen-induced arthritis (13). The specificity for B7.1 or B7.2 as the costimulatory ligand in the development of autoimmunity is not as clear. Anti-B7.1 inhibited the development of EAE, while anti-B7.2 exacerbated disease (14, 15, 16). In contrast, anti-B7.2 suppressed and anti-B7.1 accelerated the development of diabetes in NOD mice (11), while neither anti-B7.1 nor anti-B7.2 alone suppressed or accelerated the development of collagen-induced arthritis (17) or autoimmunity in NZB/NZW mice (18).
The ability to induce severe EAT with granulomatous histopathology by the adoptive transfer of MTg-primed spleen cells activated in vitro with MTg and IL-12, makes this an interesting model of autoimmunity for examining the roles of B7.1 and/or B7.2 in the activation and effector stages of autoreactive T cells. Surprisingly, a differential effect of blocking B7.2 was observed dependent upon the stage of EAT induction. Blocking B7.2 either during in vivo priming of donor mice or during in vitro activation of EAT effector cells suppressed the ability of MTg-primed spleen cells to transfer granulomatous EAT. However, treatment of recipient mice with anti-B7.2 increased EAT severity and granulomatous histopathology, suggesting opposing roles for B7.2 in the induction of EAT. Blocking both B7.1 and B7.2 during in vivo priming, in vitro activation, or development of thyroiditis in recipient mice suppressed EAT severity, suggesting that costimulation through B7.1 or B7.2 is not only necessary for the activation of primed effector cells, but is also required during the development of thyroiditis in recipient mice.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female CBA/J mice 6–8 wk of age, obtained through Clarence Reeder at the National Institutes of Health, were used for all experiments.
Ags and immunization
MTg was prepared from pooled mouse thyroids, as previously described (2). LPS (Escherichia coli 0111:B4) was purchased from Sigma (St. Louis, MO). Donor mice were injected twice with 150 µg MTg and 15 µg LPS (i.v.) at 10-day intervals (2).
Abs and injections
The anti-B7.2-producing hybridoma, GL1, was purchased from American Type Culture Collection (ATCC; Manassas, VA) (HB-253) (19). The anti-B7.1-producing hybridoma, 16-10A1 (ATCC HB-301) was provided by Dr. Hans Reiser (20), and the anti-CTLA-4-producing hybridoma (4F10) was provided by Dr. Jeffery Bluestone (21). Anti-B7.1 and anti-B7.2 were purified from cell culture supernatants using protein G columns. Anti-B7.2 Fab fragments were generated using the papain-Fab fragment kit (Pierce, Rockford, IL). For blocking B7.1 or B7.2 during in vitro activation, anti-B7.1, anti-B7.2, or a combination of anti B7.1 and anti-B7.2 was added to culture at a final concentration of 10 µg/ml of each Ab. Rat (Sigma) or hamster Ig (Jackson ImmunoResearch, West Grove, PA) (10 µg/ml) were used as controls. For blocking B7.1 and/or B7.2 during in vivo priming, 100 µg of anti-B7.1, anti-B7.2, or rat and hamster Ig was injected i.p. to donor mice on days -1, 0, and +1 relative to each immunization with MTg and LPS. For treatment of recipient mice, 100 µg of anti-B7.1, anti-B7.2, or both Abs was injected i.p. every 3 days starting on the day of spleen cell transfer, unless otherwise indicated. Control mice received 100 µg of rat and/or hamster Ig according to the same schedule.
Adoptive transfer of EAT
Seven days after the second immunization with MTg and LPS, donor spleen cells were suspended at 107 cells/ml in RPMI 1640 containing 25 mM HEPES buffer, 5% FBS (Sigma; Lot 504-0269), and 1% of the following: glutamine, vitamins, nonessential amino acids, sodium pyruvate, antibiotics, and 0.005 M 2-ME. Cells were cultured in 60-mm petri dishes (Corning 25010; Corning Glass, Corning, NY) at 4 ml/plate with 25 µg/ml MTg, anti-IL-2R mAb (M7/20) (22) (5% final concentration of culture supernatant), and 5 ng/ml IL-12 (R&D Systems, Minneapolis, MN) (5). After 72 h at 37°C, cells were harvested and washed, and 3 x 107 cells were injected i.v. into irradiated (600 rad) recipient mice (4).
Proliferation assay
For some experiments, proliferative responses of sensitized
donor spleen cells were assessed, as previously described (23). Con A
(0.5 µg/well), MTg (5 µg/well), or media were added to triplicate
wells containing 5 x 105 spleen cells. After 72
h, 0.5 µCi [3H]TdR was added to each well, and cells
were harvested 16–20 h later. Results are expressed as the mean cpm of
triplicate wells minus the mean cpm of cells cultured with medium alone
(
cpm).
ELISA assays
MTg-specific serum IgG Abs were determined by ELISA using MTg-coated ELISA plates. Serial dilutions of serum from individual mice were tested in duplicate, as previously described (24). Alkaline phosphatase-conjugated Abs specific for IgG1, IgG2A, and IgG2B (Zymed, San Francisco, CA) were used to assess the contributions of various IgG subclasses to the total IgG autoantibody response. Secondary Abs were titrated and used at an optimal concentration to provide maximal readings on MTg-coated ELISA plates and OD405 < 0.05 on OVA-coated ELISA plates.
Evaluation of EAT
Thyroids were removed from recipient mice 19 to 21 days after cell transfer, the time of peak disease severity in this model (4). EAT severity (the extent of destruction of thyroid follicles) was scored on a scale of 1+ to 5+, as previously described (4, 5). A 1+ EAT lesion is defined as an infiltrate of at least 125 cells in one or several foci; 2+ is 10–20 foci of cellular infiltrate involving up to 1/4 of the gland; 3+ indicates 1/4 to 1/2 of the gland is infiltrated; 4+ indicates that greater than 1/2 of the gland is infiltrated; and 5+ indicates that the entire thyroid is almost completely destroyed, with few or no remaining follicles.
Thyroid lesions were also evaluated qualitatively (5). The inflammatory infiltrate in conventional (lymphocytic) EAT consists primarily of lymphocytes and some plasma cells, with relatively few neutrophils. Thyroids designated "G" in tables had granulomatous lesions, characterized by proliferation and enlargement of thyroid follicular cells with lymphocytes, histiocytes, multinucleated giant cells, and small to moderate numbers of neutrophils. The more severe (4–5+) granulomatous lesions had intense inflammation dominated by neutrophils, which also included microabscess formation, necrosis, fibrin deposition, multinucleated giant cells, numerous histiocytes, and focal fibrosis, which predominated over the follicular cell proliferative changes and mononuclear infiltrates characteristic of the milder granulomatous lesions. The granulomatous inflammation in thyroids with 4–5+ severity characteristically extended beyond the thyroid to involve the adjacent connective tissue and muscle.
Statistical analysis
All p values were determined using a two-tailed
Student’s t test, in which each experimental group was
compared individually with the isotype control group. Values of
p 
0.05 were designated by *, and values of
p 0.01 were designated by **.
PCR analysis
Total RNA was isolated from 6 x 106 in vitro
activated spleen cells or from individual thyroid lobes using Trizol
(Life Technologies, Gaithersburg, MD). Total RNA was converted to cDNA
by Moloney murine leukemia virus reverse transcriptase (Perkin-Elmer
Cetus, Branchburg, NJ) and oligo(dT)12–18 primers. To
determine the relative amounts of target cDNA, 1/5 and 1/25 dilutions
of each cDNA sample were analyzed. Cytokine message was determined
using specific primers for IL-2, IL-4, IL-10, IL-12, and IFN-
,
as previously described (24, 25). B7.1 mRNA was detected using the
forward primer AGTTGTCCATCAAAGCTGAC and reverse primer
CTAAAGGAAGACGGTCT (26). B7.2 mRNA was detected using the forward primer
ATGGACCCCAGATGCACCAT and reverse primer TTTCCTCAGGCTCTCACTGC (26). HPRT
was used as a housekeeping gene to verify that similar amounts of RNA
were amplified. Samples were amplified at 28 cycles for IFN (and HPRT),
34 cycles for IL-12 and IL-10, and 42 cycles for IL-2 and IL-4. PCR
products were separated by gel electrophoresis, stained with ethidium
bromide, and analyzed on a digital imager. Cytokine message level was
determined as a ratio of cytokine message versus HPRT message for each
sample.
Cytokine ELISA
Supernatants were collected from 72-h spleen cell cultures.
IL-2, IL-4, IL-10, and IFN- levels were measured using sandwich
ELISA. JES5-2A5 (PharMingen, San Diego, CA) and biotinylated SXC-1
(PharMingen) were used to measure IL-10. 11B11 (ATCC HB-188) and
BVD6-24G2 (PharMingen) were used to measure IL-4, while
biotinylated XMG1.2 and R4-6A2 (ATCC HB170) were used to measure
IFN-. IL-2 (Endogen, Cambridge, MA) ELISA kits were used to
determine IL-2 levels.
Immunohistochemistry
Immunohistochemistry was performed as previously described (27). Thyroid lobes were removed from recipient mice 19 days after cell transfer, snap frozen in OCT (VWR Scientific, Torrance, CA) with liquid nitrogen, and stored at -70°C before cutting. Cryostat sections (6 µm) were cut, fixed in acetone for 10 min, and stored at -70°C. Sections were blocked with 1% BSA in PBS before incubation with primary Ab. Unconjugated 16-10A1 and GL1 were used to detect B7.1 and B7.2, respectively, while rat Ig (Sigma) and hamster Ig (Jackson ImmunoResearch) were used to control for nonspecific binding. Biotinylated goat anti-hamster or goat anti-rat IgG (Caltag, Burlingame, CA) were used to detect anti-B7.1 and anti-B7.2, respectively. Sections were then incubated with 0.3% H2O2, followed by avidin-biotin-horseradish peroxidase (Vectastain Elite ABC; Vector, Burlingame, CA) and VIP peroxidase substrate (Vector). To determine whether the anti-B7.1 and anti-B7.2 given to recipients bound to B7.1 and B7.2 molecules in the thyroid, frozen thyroid sections from recipient mice given anti-B7.1 and/or anti-B7.2 were incubated with anti-hamster Ig or anti-rat Ig to detect anti-B7.1 and anti-B7.2 Ab, respectively. Slides were then incubated with 0.3% H2O2, avidin-biotin-horseradish peroxidase, and VIP substrate to detect positive cells.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To determine the concentration of Ab necessary to block B7.1 or
B7.2 in vitro, 5–50 µg/ml of anti-B7.1 or anti-B7.2 was
added to MTg-primed spleen cells. A total of 10 µg/ml of each Ab
resulted in maximal suppression of MTg-specific proliferation (data not
shown). Anti-B7.1 suppressed MTg-specific proliferation approximately
twofold, while anti-B7.2 suppressed proliferation to near
background levels (Table I
). To determine
the costimulation requirements during in vitro activation of
MTg-specific effector cells, anti-B7.1 and/or anti-B7.2 were
added to MTg-primed spleen cell cultures at a final concentration of 10
µg/ml. MTg-primed spleen cells cultured with anti-B7.2 or
anti-B7.2 Fab transferred less severe EAT than spleen cells
cultured with rat Ig (Table I, Expts. 1 and 2), indicating a
requirement for B7.2 costimulation during in vitro activation of EAT
effector cells. EAT induced by spleen cells cultured in vitro with
anti-B7.1 was generally similar to that induced by cells cultured
with rat Ig (Table I, Expt. 1). To determine whether B7.2 was also
necessary for in vivo priming of effector cells, 100 µg of
anti-B7.2 was given i.p. to donor mice on days -1, 0, and +1
relative to immunization with MTg and LPS. Recipients receiving cells
from anti-B7.2-treated donor mice had suppressed EAT severity
compared with controls given rat Ig (Table I, Expts. 3 and 4),
indicating that B7.2 costimulation was also necessary for the priming
of effector cells in vivo. Spleen cells from anti-B7.2-treated
donors or spleen cells cultured with anti-B7.2 also had decreased
ability to transfer lymphocytic EAT compared with spleen cells from rat
Ig-treated controls (data not shown). Blocking both B7.1 and B7.2
during either in vivo priming or in vitro activation of MTg-primed
effector cells suppressed the transfer of EAT to a similar extent as
anti-B7.2 alone, while blocking B7.1 alone did not consistently
affect lymphocytic or granulomatous EAT severity (Table I, data not
shown).
|
), IgG1, and
IgG2A (Fig. 1) autoantibodies than
recipients of cells cultured with rat Ig. Suppression of anti-MTg
IgG1 and IgG2A responses in recipients of cells cultured with
anti-B7.1 was observed only at high serum dilutions (Fig. 1). The
effects of blocking B7.1 and/or B7.2 in vitro on anti-MTg Ab
responses were not subclass specific, since IgG1 and IgG2A (Fig. 1)
responses were suppressed to a similar extent. Blocking B7.2 during in
vivo priming suppressed both donor (data not shown) and recipient
(Table I, Expts. 3 and 4) anti-MTg autoantibody responses.
|
Culture supernatants from spleen cells cultured with either
anti-B7.2 alone or both anti-B7.1 and anti-B7.2 had
decreased IL-2 compared with supernatants from cells cultured with rat
Ig, while IL-2 production by spleen cells cultured with anti-B7.1
was not consistently decreased (Fig. 2).
IFN (and IL-10) levels were not decreased in supernatants of cells
cultured with anti-B7.1, anti-B7.2, or both anti-B7.1 and
anti-B7.2. IFN levels were also not decreased by anti-B7.1
and/or anti-B7.2 when cells were cultured in the absence of IL-12.
IL-4 was undetectable in any of the culture supernatants at 24, 48, or
72 h. Spleen cells cultured with anti-B7.2 had decreased IL-4
mRNA, as determined by RT-PCR, compared with spleen cells cultured with
rat Ig (data not shown).
|
B7 expression on thyroid infiltrating cells, but not on
thyrocytes, was observed in patients with Hashimoto’s thyroiditis or
Graves disease (28). The infiltrating cells in the target organs in
other human autoimmune diseases (29) and experimental models of
autoimmunity (15, 30) also expressed high levels of B7.1 and/or B7.2.
Immunohistochemistry analysis of thyroids with 3–5+ EAT established
the expression of both B7.1 and B7.2 in EAT thyroids, but not in normal
thyroids (Fig. 3, A and
B, data not shown). The number of B7.2+ cells
(Fig. 3B) appeared to be higher than the number of
B7.1+ cells (Fig. 3A), although it is possible
that the anti-B7.2 Ab has a higher affinity for its Ag and is
therefore more readily detectable. To examine the kinetics of B7.1 and
B7.2 expression, thyroids were removed from recipient mice at various
times after cell transfer and analyzed for B7.1 and B7.2 mRNA
expression by RT-PCR. Although B7.1 and B7.2 mRNA were undetectable in
normal thyroids, both B7.1 and B7.2 mRNA were detected as early as 5 to
7 days after cell transfer (Fig. 4). The
increased expression of B7.1 and B7.2 mRNA was maintained through peak
severity of EAT at day 21. These results demonstrate that both B7.1 and
B7.2 mRNA are expressed in autoimmune thyroid lesions and therefore may
be involved in the development of EAT in recipient mice.
|
|
To determine whether B7.1 and/or B7.2 costimulation was necessary
for the development of thyroid lesions, 100 µg of anti-B7.1 or
anti-B7.2, or both anti-B7.1 and anti-B7.2 was given to
recipient mice every 3 days beginning on the day of cell transfer.
Anti-B7.1 (hamster Ab) was detected using anti-hamster Ig (Fig. 3C), but not anti-rat Ig (Fig. 3D), in
thyroid sections from anti-B7.1-treated recipient mice. Conversely,
anti-B7.2 (rat Ab) was detected using anti-rat Ig (Fig. 3F), but not anti-hamster Ig (Fig. 3E), in
thyroid sections from anti-B7.2-treated recipient mice, while both
anti-B7.1 and anti-B7.2 were detected in thyroid sections from
recipients given both anti-B7.1 (Fig. 3G) and
anti-B7.2 (Fig. 3H). This demonstrates that both
anti-B7.1 and anti-B7.2 Abs can gain access to the thyroid
after i.p. injection, indicating that the B7.1 and/or B7.2 molecules in
the thyroid can be blocked by anti-B7.1 and/or anti-B7.2,
respectively.
Recipient mice receiving both anti-B7.1 and anti-B7.2 had
decreased EAT severity compared with recipients receiving control rat
and hamster Ig (Table II; Fig. 5, A and D),
indicating that T cell costimulation was necessary for the development
of maximal EAT in recipient mice. Although treatment of recipient mice
with anti-B7.1 suppressed the production of anti-MTg
autoantibody, it had little effect on EAT severity or the development
of granulomatous lesions (Table II, Fig. 5B). Surprisingly,
recipients treated with anti-B7.2 consistently had more severe EAT
than recipients receiving control rat and hamster Ig (Table II, Fig. 5C), suggesting a possible regulatory role for B7.2 during
the effector stage of EAT. Treatment of recipients with anti-B7.2
Fab also resulted in increased EAT severity (Table II, Expt. 4),
indicating that the increase in EAT severity was not due to
cross-linking and stimulation of recipient cells by the anti-B7.2
Ab. Thyroids from anti-B7.2 or anti-B7.2 Fab-treated recipients
had more fibrosis, follicular proliferation, and fibrin deposition,
with increased neutrophils and histiocytes and fewer lymphocytes
compared with thyroids from recipients receiving rat Ig (Fig. 5, A versus C). The lack of suppression of EAT
severity by either anti-B7.1 or anti-B7.2 alone suggests that
either costimulatory molecule can be utilized for EAT development. The
suppression of autoantibody by anti-B7.1, or anti-B7.1 and
anti-B7.2 was observed for both IgG1 and IgG2A (Table II).
|
|
The suppression of EAT by treatment of recipient mice with
anti-B7.1 and anti-B7.2 could be due to blocking costimulation
of effector cells in the periphery of recipient mice before
infiltration into the thyroid or to suppression of effector cell
costimulation in the thyroid, since both B7.1 and B7.2 are blocked in
the thyroid (Fig. 3C–H). To determine when B7.1 and B7.2
signaling was needed during the development of thyroiditis,
anti-B7.1 and anti-B7.2 or anti-B7.2 alone were
administered to recipient mice beginning on day 0 until the removal of
thyroids on day 19, as in the previous experiments, or beginning on day
7, when infiltrating cells are readily detectable in the thyroids of
recipient mice, until day 19. Mice receiving anti-B7.1 and
anti-B7.2 from day 0 to day 19 had decreased EAT severity compared
with controls, while delaying treatment with anti-B7.1 and
anti-B7.2 until day 7 (day 7 to 19) had no effect on EAT severity
(Table III). These results suggest that
the suppression of EAT by the combination of anti-B7.1 and
anti-B7.2 is most likely due to blocking costimulation either
before or during the initial infiltration of effector cells into the
thyroid. The ability of anti-B7.2 to increase EAT severity was
independent of when anti-B7.2 was given to recipients, since
anti-B7.2 increased EAT severity when given from day 0 to day 19 or
from day 7 to day 19 (Table III). This suggests that the enhancement of
EAT by blocking B7.2 can occur in the thyroid, after the initial
infiltration of cells into the thyroid. Recipients receiving
anti-B7.2 from day 0 to day 7 also had increased EAT severity
compared with rat Ig-treated controls (data not shown). However, the
enhancement of EAT by anti-B7.2 treatment from day 0 to day 7 may
be due to the presence of residual Ab, since the in vivo
t1/2 of anti-CD40L, an Ab that also blocks
cell to cell interactions, is 12 days (31). Recipients receiving both
anti-B7.1 and anti-B7.2 from day 0 to day 7 had decreased EAT
compared with controls (data not shown), although this may also be due
to the persistence of residual Ab after day 7.
|
The suppression or enhancement of inflammation by blocking
costimulation has been associated with alterations in cytokine
production (14, 32, 33). To determine whether the effects of treatment
of recipients with anti-B7.2 or the combination of anti-B7.1
and anti-B7.2 were associated with changes in cytokine production
in the target organ, thyroids from recipient mice were examined for
mRNA expression by RT-PCR. Although intercytokine comparison is not
valid using RT-PCR due to variations in primer annealing and variable
amplification, the ratio of a particular cytokine between samples can
be semiquantitatively compared to provide some indication of
relative mRNA levels (24, 25). Thyroids were removed on day 19 after
cell transfer, and one thyroid lobe was used for histology, while the
other lobe was used for isolation of RNA (24). Thyroids from recipients
given anti-B7.1 and anti-B7.2 had decreased IL-2, IL-4, IL-13,
and IFN- mRNA expression compared with mRNA from thyroids of
recipients treated with rat Ig and hamster Ig (Fig. 6). IL-2, IL-4, IL-10, IL-12, IL-13, and
IFN- mRNA were not expressed in thyroids from naive mice, indicating
that the cytokine mRNA expression was from the infiltrating cells. A
statistically significant increase in IL-4 mRNA expression was observed
in thyroids from recipients given anti-B7.2 or anti-B7.2 Fab
fragments (Fig. 6). IL-13 (Fig. 6) and IL-10 (data not shown) mRNA
expression was increased in some, but not all, experiments. No
consistent changes in mRNA expression were observed for IL-2, IFN (Fig. 6), or IL-12 (data not shown).
|
The enhancement of EAT severity by blocking B7.2 in recipients
could be due to blocking interactions between B7.2 and CTLA-4, thereby
preventing down-regulation of activated T cells. To determine whether
blocking CTLA-4 in recipient mice would have a similar effect to
blocking B7.2, 100 µg of anti-CTLA-4 mAb was given to recipients
every 3 days beginning on the day of cell transfer. Similar to
recipients receiving anti-B7.2 (Table II), recipients given
anti-CTLA-4 had more severe EAT, with increased neutrophils and
fibrosis, than recipients receiving control hamster Ig (Table IV). These results suggest that the
enhancement of EAT by blocking B7.2 may be due to blocking B7.2-CTLA-4
interactions.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) suppressed the
ability of MTg-primed spleen cells to transfer EAT, while blocking B7.2
in recipient mice resulted in more severe granulomatous EAT (Table II).
Ag (MTg) presentation during donor priming and in vitro activation of
MTg-primed effector cells is required for the adoptive transfer of EAT
(2, 36, 37). The requirement for B7.2 and not B7.1 during in vitro
activation may be explained by the differences in B7.1 and B7.2
expression on splenic APCs, since B7.2 is prevalent on splenocytes,
while only a low level of B7.1 is expressed (38). When B7.2 is blocked,
the levels of B7.1 may be too low to allow effective T cell
costimulation. However, optimal concentrations of either B7.1- or
B7.2-deficient APCs stimulated OVA-transgenic T cells to proliferate
and produce cytokines, suggesting that B7.1 can provide costimulatory
signals in vitro when B7.2 is absent (39). Blocking both B7.1 and B7.2
during in vitro activation of effector cells suppressed the transfer of
EAE to recipient mice (16). However, neither Ab alone suppressed EAE or
proliferative responses to myelin basic protein, although
recipients of cells cultured with anti-B7.2 had less severe
relapses compared with controls (16). The differences between the
ability of anti-B7.2 to suppress adoptive transfer of EAT versus
EAE may be the source of cells activated in vitro (spleen versus lymph
node), the level of B7.1 expression on the APCs, or differences in
costimulation requirements in the two model systems. In both models,
suppression of disease correlated with a decrease in IL-2 production
during Ag activation (Fig. 2) (16). Studies are in progress to
determine whether blocking B7.2 during in vitro activation induces
anergy to MTg.
The expression of costimulatory molecules in inflammatory autoimmune
lesions has been demonstrated in autoimmune thyroiditis (28),
rheumatoid arthritis (40), and multiple sclerosis (29), as well as in
the murine models of EAE (15) and diabetes (30). Expression of B7.1 and
B7.2 in thyroids of mice with EAT, but not normal thyroids, was
observed by immunohistochemistry (Fig. 3) and by RT-PCR (Fig. 4),
suggesting that B7.1 and B7.2 are expressed on the infiltrating cells
and not on thyrocytes. The expression of B7 by infiltrating cells was
also observed in human autoimmune thyroiditis, while thyrocytes did not
express B7 (28). In multiple sclerosis, both B7.1 and B7.2 were
observed in the infiltrating cells in the central nervous system, with
a predominance of B7.1 (29). The level of B7.1 and B7.2 expression in a
particular target organ and the time of expression of B7.1 and B7.2 may
determine which of these molecules is most critical for development of
a given autoimmune disease (9). The predominant expression of B7.1 in
the central nervous system during EAE correlated with the suppressive
effects of anti-B7.1, which blocked clinical relapses of EAE,
suppressed disease severity, and prevented epitope spreading (15). In
diabetes, the predominant expression of B7.2 in the pancreas correlated
with the suppression of diabetes by anti-B7.2, but not
anti-B7.1 (41). Both B7.1 and B7.2 were expressed in thyroids from
recipient mice with EAT, with B7.2+ cells predominating
over B7.1+ cells (Fig. 3, A and B).
However, this did not correlate with the ability of anti-B7.1 or
anti-B7.2 to suppress EAT severity in recipient mice, since
anti-B7.2 treatment of recipients increased EAT severity, while
anti-B7.1 alone had no effect (Table II). The enhancement of EAT
severity by blocking B7.2 in recipients may be due to T cells receiving
costimulation only through B7.1 or by blocking interactions between
B7.2 and CTLA-4, thereby preventing down-regulation of activated T
cells. Blocking CTLA-4 after the activation of EAE effector cells (42)
or after the onset of EAE (43) increased EAE severity, while blocking
CTLA-4 in BDC2.5/NOD TCR-transgenic mice increased the severity of
spontaneous diabetes (44). The ability of anti-B7.2 to enhance EAT
severity when treatment with anti-B7.2 was delayed until 7 days
after cell transfer (Table III) suggests that anti-B7.2 may be
inhibiting B7.2-CTLA-4 interactions instead of B7.2-CD28 interactions.
This is consistent with preliminary results indicating that recipient
mice given anti-CTLA-4 have increased EAT severity and
granulomatous histopathology (Table IV) similar to that observed for
recipients given anti-B7.2 (Table II). However, if the enhancement
of EAT is due to blocking CTLA-4 signaling, it is unclear why
recipients given both anti-B7.1 and anti-B7.2, beginning on day
7 when costimulation was apparently no longer required, did not have
enhanced EAT severity (Table III).
The increase in IL-4 mRNA expression in thyroids from mice given
anti-B7.2 is surprising since other Th2 cytokines were not
increased in all experiments, and IFN was not affected (Fig. 6). In
Leishmania donovani-infected mice, treatment with
anti-B7.2 increased the number of IL-4- and IFN-producing cells
(45). Blocking CTLA-4 during a response to staphylococcal enterotoxin B
increased IL-4, but not IFN, production by CD4+ T cells
(46), suggesting that signaling through CTLA-4 may suppress Th2
differentiation. Although IL-4-deficient mice can develop granulomatous
EAT (24), administration of anti-IL-4 to CBA/J recipient mice
reduced granulomatous histopathology, but not EAT severity
(Braley-Mullen, unpublished results), suggesting that IL-4 may be
involved in promoting granulomatous histopathology in the thyroid.
The suppression of EAT by blocking B7.1 and B7.2 in recipient mice
indicates that costimulation is required during the effector stage of
EAT. Administration of anti-B7.1 Fab after the onset of clinical
signs suppressed relapses in EAE, suggesting that costimulation was
required during the effector stage of autoimmunity (15). A requirement
for costimulation after adoptive transfer was suggested when spleen
cells from diabetic NOD mice rapidly induced diabetes in transgenic
NOD.scid mice expressing B7.1 on pancreatic ß cells, but not in
nontransgenic NOD.scid mice (47). The suppression of EAT by blocking
both B7.1 and B7.2 in recipient mice is apparently due to a requirement
for costimulation of MTg-primed effector cells during early development
of thyroid lesions, since blocking B7.1 and B7.2 starting on the day of
transfer suppressed EAT, while delaying treatment until day 7 had no
effect (Table III). Treatment of recipients with both anti-B7.1 and
anti-B7.2 also suppressed the expression of IL-2, IL-4, and IFN-
mRNA in the thyroid (Fig. 6). The decreased levels of cytokine mRNA are
likely to be due to the overall decrease in inflammatory cells in the
thyroid, since the level of CD3 mRNA was also decreased in thyroids
from anti-B7.1- and anti-B7.2-treated recipients (data not
shown).
The lack of complete suppression of EAT by both anti-B7.1 and
anti-B7.2 (Table II) may be due to use of an insufficient amount of
Ab, or to a role for other costimulatory molecules/pathways in the
development of EAT. It is also possible that activation of some EAT
effector cells may be costimulation independent (48). In murine
lupus and autoimmune oophoritis, complete long-term suppression of
disease was only observed with the inhibition of both the CD28-B7 and
CD40-CD40L pathways (35, 36). However, in our studies, treatment of
recipients with a combination of anti-B7.1, anti-B7.2, and
anti-CD40L did not further suppress EAT (data not shown).
In conclusion, these results demonstrate that B7.2 has opposing roles
in the regulation of EAT. Blocking B7.2 costimulation before effector
cells are fully active (during in vivo priming or in vitro activation)
suppresses the ability of MTg-primed spleen cells to induce EAT, and
also inhibits proliferation and IL-2 production in response to MTg,
suggesting that costimulation through B7.2 is necessary for activation
of EAT effector cells (Table I). However, during the development of
thyroid lesions in recipient mice, inhibition of B7.2 increased EAT
severity (Table II), suggesting that B7.2 may be important for the
down-regulation of inflammatory responses in the thyroid. Inhibition of
both anti-B7.1 and anti-B7.2 in recipient mice resulted in
decreased EAT severity, suggesting that costimulation is necessary
during the priming, in vitro activation, and effector stages of
MTg-primed EAT effector cells. Studies are underway to determine the
role of B7.1 and/or B7.2 in the resolution of granulomatous EAT
lesions, as well as defining the role of CTLA-4 in the development and
resolution of EAT lesions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Helen Braley-Mullen, Dept. of Internal Medicine, M450 Med. Sci. Bldg., 1 Hospital Drive, University of Missouri, Columbia, MO 65212.
3 Abbreviations used in this paper: EAT, experimental autoimmune thyroiditis; CD40L, CD40 ligand; CTLA-4, cytotoxic T lymphocyte antigen-4; EAE, experimental autoimmune encephalomyelitis; HPRT, hypoxanthine-guanine phosphoribosyltransferase; MTg, mouse thyroglobulin; NOD, nonobese diabetic.
Received for publication June 19, 1998. Accepted for publication October 28, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4 integrin and intercellular adhesion molecule-1 (ICAM-1) in murine experimental autoimmune thyroiditis. Autoimmunity 23:9.[Medline]
This article has been cited by other articles:
|
|
 |
|
  N. Perez, S. Karumuthil-Melethil, R. Li, B. S. Prabhakar, M. J. Holterman, and C. Vasu Preferential Costimulation by CD80 Results in IL-10-Dependent TGF-{beta}1+-Adaptive Regulatory T Cell Generation J. Immunol., May 15, 2008; 180(10): 6566 - 6576. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Singh, P. R. Chandler, Y. Seki, B. Baban, M. Takezaki, D. J. Kahler, D. H. Munn, C. P. Larsen, A. L. Mellor, and M. Iwashima Role of CD28 in fatal autoimmune disorder in scurfy mice Blood, August 15, 2007; 110(4): 1199 - 1206. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Zhang, D. Sun, Y. Ke, H. J. Kaplan, and H. Shao The Net Effect of Costimulatory Blockers Is Dependent on the Subset and Activation Status of the Autoreactive T Cells J. Immunol., January 1, 2007; 178(1): 474 - 479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Lyddane, B. U. Gajewska, E. Santos, P. D. King, G. C. Furtado, and M. Sadelain Cutting Edge: CD28 Controls Dominant Regulatory T Cell Activity during Active Immunization J. Immunol., March 15, 2006; 176(6): 3306 - 3310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Odobasic, A. R. Kitching, T. J. Semple, J. R. Timoshanko, P. G. Tipping, and S. R. Holdsworth Glomerular Expression of CD80 and CD86 Is Required for Leukocyte Accumulation and Injury in Crescentic Glomerulonephritis J. Am. Soc. Nephrol., July 1, 2005; 16(7): 2012 - 2022. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Yadav, V. Judkowski, M. Flodstrom-Tullberg, L. Sterling, W. L. Redmond, L. Sherman, and N. Sarvetnick B7-2 (CD86) Controls the Priming of Autoreactive CD4 T Cell Response against Pancreatic Islets J. Immunol., September 15, 2004; 173(6): 3631 - 3639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Vasu, S. R. Gorla, B. S. Prabhakar, and M. J. Holterman Targeted engagement of CTLA-4 prevents autoimmune thyroiditis Int. Immunol., May 1, 2003; 15(5): 641 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Poussin, E. Tuzun, E. Goluszko, B. G. Scott, H. Yang, J. U. Franco, and P. Christadoss B7-1 Costimulatory Molecule Is Critical for the Development of Experimental Autoimmune Myasthenia Gravis J. Immunol., April 15, 2003; 170(8): 4389 - 4396. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Peng, C. Liu, D. Liu, C. Ren, W. Li, Z. Wang, N. Xing, C. Xu, X. Chen, C. Ji, et al. Effects of B7-blocking agent and/or CsA on induction of platelet-specific T-cell anergy in chronic autoimmune thrombocytopenic purpura Blood, April 1, 2003; 101(7): 2721 - 2726. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-M. Hu, H. Winter, J. Ma, M. Croft, W. J. Urba, and B. A. Fox CD28, TNF Receptor, and IL-12 Are Critical for CD4-Independent Cross-Priming of Therapeutic Antitumor CD8+ T Cells J. Immunol., November 1, 2002; 169(9): 4897 - 4904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Yamada, A. D. Salama, and M. H. Sayegh The Role of Novel T Cell Costimulatory Pathways in Autoimmunity and Transplantation J. Am. Soc. Nephrol., February 1, 2002; 13(2): 559 - 575. [Full Text] [PDF]
|
|
|
|
|
|
B. Liang, M. J. Kashgarian, A. H. Sharpe, and M. J. Mamula Autoantibody Responses and Pathology Regulated by B7-1 and B7-2 Costimulation in MRL/lpr Lupus J. Immunol., September 15, 2000; 165(6): 3436 - 3443. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Liang, R. J. Gee, M. J. Kashgarian, A. H. Sharpe, and M. J. Mamula B7 Costimulation in the Development of Lupus: Autoimmunity Arises Either in the Absence of B7.1/B7.2 or in the Presence of Anti-B7.1/B7.2 Blocking Antibodies J. Immunol., August 15, 1999; 163(4): 2322 - 2329. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), anti-B7.2 (
), both anti-B7.1 and
anti-B7.2 (•), or rat and hamster Ig (
). Serial dilutions of
serum from recipient mice (Table I
