| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
in Development of Spontaneous Autoimmune Thyroiditis in NOD.H-2h4 Mice1
,
Departments of
*
Internal Medicine and
Medical Microbiology and Immunology, University of Missouri School of Medicine, and
Department of Veterans Affairs Research Service, Columbia, MO 65212
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 mRNA
was constitutively expressed in thyroids and spleens of normal
NOD.H-2h4 mice but not other strains of mice. To determine whether
TGF might have a role in SAT, mice were given anti-TGF mAb at
various times during development of SAT. Anti-TGF markedly inhibited
development of SAT and production of anti-mouse thyroglobulin IgG1
autoantibodies. Anti-TGF was most effective in inhibiting SAT when
given during the time thyroid lesions were developing, i.e., starting 4
wk after administration of NaI water. The active form of the TGF1
protein was present in thyroids of mice with SAT but not in normal
NOD.H-2h4 thyroids. However, thyrocytes of normal NOD.H-2h4 thyroids
did express latent TGF1. TGF1 protein expression in the thyroid
correlated with SAT severity scores, and administration of
anti-TGF inhibited TGF1 protein expression in both the
thyroid and spleen. TGF1 was produced primarily by inflammatory
cells and was primarily localized in areas of the thyroid containing
clusters of CD4+ T and B cells. Depletion of
CD8+ T cells had no effect on TGF1 protein expression.
Activation of splenic T cells was apparently not inhibited by
anti-TGF, because up-regulation of mRNA for cytokines and other
T cell activation markers was similar for control and
anti-TGF-treated mice. TGF1 may function by promoting
migration to, or retention of, inflammatory cells in the
thyroid. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1,
which is expressed at relatively similar levels in thyroids and spleens
of NOD.H-2h4 mice with SAT and in normal NOD.H-2h4 mice (Ref.
4 and our unpublished results). The constitutive
expression of relatively high levels of TGF1 mRNA in spleens and
thyroids of NOD.H-2h4 mice is also observed in spleen and thyroids of
NOD.LtJ mice (our unpublished observations), but not other strains of
mice such as CBA/J and DBA/1 (10). NOD.LtJ mice
spontaneously develop type 1 diabetes, and often have inflammatory cell
infiltrates in other organs, including the thyroid (11, 12).
TGF1 is a pleiotropic cytokine known to play an important role in
regulating immune responses and various infectious and autoimmune
diseases (13, 14, 15). TGF can have both positive and
negative effects, depending on the state of differentiation of the
cells and the cytokine milieu (13, 14, 15, 16). In general, TGF
suppresses autoimmune diseases when administered in vivo (13, 17, 18, 19, 20, 21, 22), and mice lacking TGF1 due to gene deletion die at
3–4 wk of age from massive lymphoproliferation and autoimmunity
(14). However, effector cells activated with Ag and
TGF1 in vitro can transfer more severe experimental allergic
encephalomyelitis (23) or experimental autoimmune
thyroiditis (EAT) (24) to recipient mice and
autoimmune-prone MRL/lpr mice, and SLE patients can produce
increased levels of active TGF1 during disease exacerbations
(25, 26, 27). TGF can be expressed at relatively high
levels in target tissues of animals and humans with autoimmune diseases
(28, 29, 30, 31) and may play a role in promoting the migration to
and accumulation of cells at the inflammatory site (31).
These latter observations (25, 26, 27, 28, 29, 30, 31) seem to be consistent
with our observation that NOD.H-2h4 mice constitutively express high
levels of TGF1 mRNA in their spleens and thyroids, and raise the
possibility that local production of TGF1 could be important for
development of SAT. To address this possibility, the current study was
undertaken to determine whether neutralization of TGF would alter
the development of SAT in NOD.H-2h4 mice and to determine whether
protein expression of TGF1 in the thyroid was increased before
and/or during development of SAT. The results indicate that
neutralization of TGF profoundly inhibits development of SAT in
NOD.H-2h4 mice. The production of the active form of TGF1 protein in
the thyroid correlates with the severity of thyroid lesions, suggesting
that TGF1 produced in the thyroid may be important for development
of SAT in NOD.H-2h4 mice.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NOD.H-2h4 mice, derived by crossing NOD mice with B10.A(4R) with repetitive backcrosses to NOD using progeny expressing the MHC haplotype of B10.A(4R), were provided by Dr. L. Wicker (Merck Laboratories, Rahway, NJ). The mice were subsequently bred and maintained under specific pathogen-free conditions in the animal facilities at the University of Missouri (Columbia, MO) (4). Both male and female mice were used for these experiments, and mice were age (±2 wk)- and sex-matched for each individual experiment. All mice received 0.05% NaI in their drinking water beginning at 7–8 wk of age (4), and thyroids were removed 8–9 wk later (when SAT reaches maximal severity) (4) unless indicated otherwise.
Ab and treatment of mice
The cell line producing the anti-TGF1 mAb 1D11.16.8
(mouse IgG1) was obtained from the American Type Culture Collection
(Manassas, VA) (HB9849) (32) and purified from cell
culture supernatants using protein G-Sepharose columns (Amersham
Pharmacia Biotech, Piscataway, NJ). Mice were given 250 µg of mAb
i.p. as indicated in each table. The amount and schedule of injections
of 1D11.16.8 was determined to be effective in preliminary experiments
and was based on results of others using this mAb in MRL mice
(25) and in a rat model of arthritis (30).
Age- and sex-matched controls were given the same amount of normal
mouse 
globulin (Jackson ImmunoResearch Laboratories, West Grove,
PA) or saline. Preliminary experiments established that identical
results were obtained whether control mice were injected with saline or
with mouse globulin. In some experiments, mice were given three
injections of anti-CD8 mAb (ATCC HB129) at 10-day intervals
beginning 5 wk after NaI water. As shown previously (4),
depletion of CD8+ T cells before development of
thyroid lesions inhibits SAT development. However, if
CD8+ T cells are depleted later, after SAT
lesions begin to develop, CD8+ T cells are almost
completely depleted from thyroids but there is no effect on SAT
severity scores.
Assessment of thyroiditis
Thyroids were collected, and one thyroid lobe from each animal was fixed in formalin, sectioned, and stained with H&E as previously described (4). The other thyroid lobe from each animal was snap-frozen in liquid nitrogen and stored at -70°C for analysis by immunohistochemistry or RT-PCR (see below). As shown previously (4), thyroid lesions reach maximal severity 7–8 wk after mice are given 0.05% NaI in their drinking water beginning at 2 mo of age. Thyroid lesions are chronic and remain relatively unchanged in severity for at least the next 14 wk (our unpublished observations). H&E-stained thyroid sections were scored for the extent of follicle destruction (SAT severity score) using a scale of 1+ to 4+ as previously described (4, 8). Briefly, 1+ thyroiditis is defined as an infiltrate of at least 125 cells in one or several foci; 2+ represents 10–20 foci of cellular infiltration, each the size of several follicles and involving up to 25% of the gland; 3+ indicates that 25–50% of the gland is destroyed by infiltrating inflammatory cells; and 4+ indicates that >50% of the gland is destroyed. Qualitatively, the inflammatory cell infiltrate was typical of that seen in conventional lymphocytic EAT, consisting primarily of lymphocytes and other mononuclear cells (4, 33), although thyroids of mice with SAT have many more B cells and fewer CD8+ T cells than mice with EAT (9). Some thyroids also had some proliferation and enlargement of thyroid follicular cells. Follicular cell enlargement was also evident in many thyroids with insufficient inflammatory cell infiltration to receive a severity score of 1+ and was probably a consequence of the increased dietary iodine. All slides were coded before being scored by two individuals, one of whom had no knowledge of the experimental protocol.
Autoantibody determination
Mouse thyroglobulin (MTg)-specific IgG1 and IgG2b autoantibodies were determined by ELISA using serum from individual mice as previously described (4). The secondary Abs were used at previously determined optimal dilutions (1/6000 or 1/8000) that gave an OD of <0.05 with a 1/50 dilution of normal mouse serum (from young NOD.H-2h4 mice not given NaI water) on plates coated with MTg. All normal mouse sera and test sera were also tested on plates coated with an irrelevant protein (OVA) and always gave an OD <0.05. Results are expressed as OD410 of 1/50 and 1/200 dilutions of serum. All assays were repeated at least once and sera from several experiments were assayed on the same day to correct for day-to-day variation in OD values.
RT-PCR
RNA was isolated from individual thyroid lobes or spleen
fragments using TRIzol (Life Technologies, Gaithersburg, MD) as
previously described in detail (4, 10). Total mRNA was
converted to cDNA, and each sample was serially diluted 1/5, 1/25, and
1/125 and amplified with specific primers. Hypoxanthine phosphoribosyl
transferase (HPRT) was used as a housekeeping gene to verify that the
same amount of RNA was amplified. All primer sequences have been
previously described (10, 34). To compare relative amounts
of mRNA transcripts between different groups, samples were reverse
transcribed and amplified at the same time using aliquots of reagent
from the same master mix. PCR products were collected before
amplifications reached the plateau phase, separated by electrophoresis,
and visualized by UV light following ethidium bromide staining.
Densitometry analysis was performed using the IS-1000 Digital Imaging
System (Life Sciences, St. Louis, MO). Samples within the linear
relationship between input cDNA and final PCR products (usually 1/25
cDNA dilutions) were used for analysis, and the densitometric units for
each cytokine band were normalized to those of the corresponding HPRT
bands (10, 34). The resulting values were multiplied by
100 for generating the values shown in Tables IV
and V, i.e., a ratio
of cytokine:HPRT of 50 indicates the HPRT band is 2-fold higher than
the corresponding cytokine band.
|
|
Production of the active form of TGF1 protein (35, 36) in thyroids and spleens of NOD.H-2h4 mice was determined
using immunohistochemical staining as previously described in detail
(37). Briefly, tissue sections were deparaffinized in
xylene, rehydrated in ethanol, and rinsed in PBS. Endogenous peroxidase
activity was blocked with 0.3% hydrogen peroxide. Tissues were
pretreated with 20 µg/ml proteinase K for 20 min and blocked with 5%
normal donkey serum (Jackson ImmunoResearch Laboratories). Sections
were then incubated for 1 h with polyclonal chicken
anti-TGF1, diluted 1/100 (AB101-NA; R&D Systems, Minneapolis,
MN). Biotin-conjugated donkey anti-chicken Ab (Jackson
ImmunoResearch Laboratories) diluted 1/500 was used as the secondary
Ab. This was followed by incubation with ABC (Vector
Laboratories, Burlingame, CA) for 30 min. Diaminobenzidine
tetrahydrochloride (Sigma-Aldrich, St. Louis, MO) was used as the
chromogen, and sections were counterstained with hematoxylin. The
specificity of staining was confirmed each time by omission of the
primary Ab or by substitution of nonimmune chicken IgY in place of
primary Ab. These controls were always negative. Paraffin and frozen
sections of thyroids were stained for latent TGF (35, 36) using anti-human latency-associated peptide (LAP)
(AB-246-NA; R&D Systems). After blocking with 1% BSA and 0.3%
hydrogen peroxide, slides were incubated overnight at 4°C with
primary Ab diluted 1/100 (paraffin sections) or 1/300 (frozen
sections). Following incubation with a secondary biotinylated rabbit
anti-goat IgG (Jackson ImmunoResearch Laboratories) diluted 1/500,
immunoreactivity was detected using ABC with diaminobenzidine
tetrahydrochloride as the chromogen. Controls with primary Ab replaced
by goat IgG were always negative. Staining of paraffin sections for
B220 was conducted in a similar manner using rat anti-mouse B220
(Caltag Laboratories, Burlingame, CA) diluted 1/500 as the primary Ab
and biotinylated goat anti-rat IgG (Caltag Laboratories) diluted
1/500 as the secondary Ab.
Statistical analysis
A two-tailed Student’s t test was used to determine
the significance of differences in SAT severity scores and cytokine
mRNA between different groups. Values of p <0.05 were
considered to be significant. The p values are given in the
footnotes in Tables I and II. Significant differences are indicated by
* in Tables IV
and V.
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As noted in the introduction, spleens and thyroids of normal
NOD.H-2h4 mice constitutively express levels of TGF1 mRNA comparable
to those of mice with SAT (Ref. 4 and our unpublished
results). To determine whether TGF1 might be important for
development of SAT, NOD.H-2h4 mice were given four weekly injections of
250 µg of anti-TGF mAb (1D11.16) or normal mouse Ig beginning
at 12 wk of age, 4 wk after being given NaI in the water. Previous
studies (4, 5, 6) have shown that thyroid lesions are
beginning to develop in most mice at this time. As shown in two
representative experiments in Table I, mice given anti-TGF mAb
had a much lower incidence and severity of SAT than did controls given
mouse Ig. Anti-MTg autoantibody responses, particularly those of the
IgG1 isotype, were also lower for most anti-TGF-treated mice
compared with controls, while IgG2B responses were only slightly lower
than in controls. Further titration of the sera (1/400, 1/800) did not
reveal greater differences in IgG2b responses for controls vs
anti-TGF-treated mice (data not shown). As described previously
(4), IgG2A, IgG3, and IgA anti-MTg autoantibodies were
not detected in sera of these mice. The fact that IgG1 autoantibody
responses were generally decreased in anti-TGF-treated mice
relative to controls, while IgG2B responses were not, was surprising
because other studies have suggested that TGF is important for IgG2B
responses (38). We currently have no explanation for this
result, which was observed in most of these experiments.
Anti-TGF suppresses SAT when given during development of
thyroid lesions, but the same amount of mAb does not reverse
established SAT or suppress SAT when given early in life
To begin to address the possible mechanisms by which TGF might
influence SAT development, mice were given anti-TGF at various
times during and after development of thyroid lesions. In the first
experiment shown in Table II, groups of
mice were given weekly injections of anti-TGF beginning 2, 4, 6,
or 8 wk after receiving NaI in the water. In this experiment, two
injections of anti-TGF mAb beginning 6 wk after NaI water (Table II, line 4) suppressed SAT as effectively as when mAb injections were
begun at 2 (Table II, line 2) or 4 (Table II, line 3) wk. In several
similar experiments, delaying mAb injections until 6 wk after NaI water
(when thyroid lesions have developed but had not reached maximal
severity (4)) always suppressed SAT but was sometimes less
effective than when mAb injections were begun at 4 wk (e.g., Table II,
line 9 vs line 10). When anti-TGF was given to mice beginning at
8 wk, when thyroid lesions were maximal, there was no effect on the
severity of SAT 3 wk later (Table II, line 5 vs line 6). In several
similar experiments, two injections of anti-TGF starting 8–9 wk
after NaI water partially reversed established SAT in only one
experiment (data not shown). The inability of anti-TGF to
reverse established SAT may be due to incomplete neutralization of
TGF, because expression of TGF1 protein in the thyroid is maximal
when SAT severity scores are maximal (see below). The amount of
anti-TGF used for these experiments (250 µg/wk for 2 or 4 wk)
effectively inhibited TGF1 protein in spleens and thyroids when mAb
was administered before 8 wk (Fig. 1, E and H, and data not shown) but only partially
inhibited TGF1 protein in the thyroid when administered at 8 and 9
wk (Fig. 1G). Therefore, suppression of established SAT may
require higher amounts of anti-TGF sufficient to neutralize most
of the TGF1 in the thyroid. Experiments to address this possibility
are in progress.
|
often suppresses some autoimmune diseases (13, 17, 18, 19, 20, 21, 22), it was possible that SAT might develop earlier and/or
be more severe if TGF was neutralized during the first few weeks
after birth. In the second experiment in Table II, groups of mice were
given four weekly injections of 250 µg of anti-TGF or mouse Ig
beginning at 3 or 12 wk of age, and another group received two
injections of anti-TGF beginning at 14 wk of age. All mice were
given NaI in the water at 8 wk of age, and thyroids were removed 8 wk
later for determination of SAT severity. As in previous experiments,
mice given anti-TGF beginning at 12 or 14 wk of age (4 or 6 wk
after NaI water) developed less severe SAT than controls. Early
administration of anti-TGF (beginning at 3 wk) had no
significant effect on SAT severity (line 8 vs line 7), and three of the
five mice developed 2+ to 3+ SAT. Although the other two mice
developed minimal SAT (thyroids had insufficient inflammation to
receive a severity score of 1+), this was also true for one of the six
mouse Ig-treated controls. In other experiments (not shown),
anti-TGF given to mice at various times ranging from shortly
after birth to 7 wk of age had no consistent enhancing or suppressive
effects on development of SAT.
Expression of TGF1 protein in thyroids of mice with SAT
The above results suggest TGF1 is important for development of
SAT. As TGF1 mRNA is constitutively expressed in thyroids of
NOD.H-2h4 mice (4), it was important to determine whether
TGF1 protein was produced in NOD.H-2h4 thyroids. Expression of
TGF1 protein in thyroids from both normal NOD.H-2h4 mice (no NaI
water and no SAT) and from NOD.H-2h4 mice with various SAT severity
scores was determined by immunohistochemical staining as described in
Methods using an anti-TGF Ab that recognizes the active form of
TGF (35, 36, 37). Although TGF1 mRNA was highly
expressed in thyroids of normal NOD.H-2h4 mice (4), the
active form of the TGF1 protein was detected only in thyroids of
mice with SAT (Fig. 1, A–C). The level of
expression of TGF1 protein in the thyroid generally correlated with
the SAT severity score, with many inflammatory cells in thyroids with
3+ severity scores producing high levels of TGF1 protein (Fig. 1, C, E, and F, and Table III). Most of the TGF1 protein
appeared to be produced by infiltrating inflammatory cells (Fig. 1E), although a few thyroid follicular cells in some
thyroids with 2+ to 3+ severity scores were also positive for TGF1
(Fig. 1, C, F, and K, arrows). Spleens
of NOD.H-2h4 mice with SAT also expressed the active form of TGF1
protein, whereas spleens of normal NOD.H-2h4 mice (no SAT) and mice
given anti-TGF before 8 wk had nearly undetectable TGF1
protein (Fig. 1, H and I, and Table III). In
general, TGF1 protein expression in the spleens of NOD.H-2h4 mice
with SAT was low compared with that observed for thyroids from the same
animals (Table III). Because the chicken anti-TGF1 Ab used for
staining primarily recognizes an active form of TGF1 as described by
the manufacturers and by others (35, 36, 37), the TGF1
detected in spleens and thyroids of NOD.H-2h4 mice with SAT is probably
the active form.
|
1 mRNA was highly expressed in thyroids and spleens
of normal NOD.H-2h4 mice, whereas TGF1 protein was not, could be
explained if the latent form of TGF1 protein was expressed in normal
spleens and thyroids. To address this possibility, thyroids of normal
NOD.H-2h4 mice and thyroids of mice with SAT were stained using an Ab
specific for the LAP of TGF1 (35, 36). Many of the
thyroid follicular cells of normal NOD.H-2h4 mice, as well as thyroid
follicular cells and some infiltrating inflammatory cells from
NOD.H-2h4 mice with SAT, expressed latent TGF1 protein (Fig. 1, M and N); latent TGF was nearly undetectable
on thyrocytes of normal CBA/J mice (data not shown). These results
suggest that expression of latent TGF1 protein could explain the
constitutive expression of TGF1 mRNA expression in NOD.H-2h4 mice.
We have not yet determined whether spleen cells also express latent
TGF.
Administration of anti-TGF under conditions that
resulted in inhibition of SAT reduced TGF1 protein expression in
both the spleen and thyroid (Table III and Fig. 1, D and
I). However, as noted above, the same amount of
anti-TGF only partially decreased TGF1 protein when mAb was
administered beginning at 8 wk, when SAT severity and TGF1 protein
expression in the thyroid were maximal (Fig. 1, F vs
G, Tables II and III). Administration of anti-TGF
generally had little effect on TGF1 mRNA expression in either the
spleen (data not shown) or thyroid (Table IV); whether anti-TGF modulates
the expression of latent TGF protein has not been determined.
Expression of active TGF1 protein in thyroids was not affected by
depletion of CD8+ T cells under conditions that
almost completely eliminated CD8+ T cells in the
thyroid but had no effect on SAT severity scores (4) (Fig. 1J and Table III). This suggests that
CD8+ T cells are not a primary source of active
TGF1 in NOD.H-2h4 thyroids, although dual staining will be needed to
directly show that CD8+ cells do not produce
TGF1 in NOD.H-2h4 thyroids. Active TGF1 protein produced by
inflammatory cells was often localized in areas of the thyroid shown
previously (9) to contain clusters of
CD4+ T cells and B220+ B
cells (Fig. 1, K and L). Because
CD4+ T and B cells are very closely clustered
together in SAT thyroids (9), it may not be possible to
determine whether only one or both of these cell types are the primary
producers of TGF1. Because depletion of either B cells
(8) or CD4+ T cells (4)
almost completely inhibits SAT, depletion of either of these cell types
would probably result in reduced TGF1 protein in the thyroid because
TGF1 protein expression in thyroids correlates with SAT severity
scores. Dual staining for CD4 and TGF1 or B220 and TGF1 should
allow us to address this issue; these studies are in progress.
Effect of administration of anti-TGF on cytokine
gene expression in spleens and thyroids
Our previous studies showed that thyroids of NOD.H-2h4
mice with SAT, but not thyroids of normal mice, express mRNA for both
Th1 and Th2 cytokines, and cytokine mRNA levels generally correlate
with the SAT severity score, which is an indication of the number of
infiltrating inflammatory cells (4). Thyroids of
anti-TGF-treated mice had lower expression of IL-2, IL-4, and
IFN- mRNA compared with thyroids of control mice (Table IV), as
would be expected because thyroids of anti-TGF-treated mice had
fewer infiltrating inflammatory cells (0 to 1+ severity scores) than
those of most control mice (2+ to 3+ severity scores). Expression of
TGF1 mRNA was comparable for control vs anti-TGF-treated mice
as well as normal mice.
To determine whether activation of peripheral autoreactive
T cells was inhibited by anti-TGF, mRNA was isolated from
spleens of control and anti-TGF-treated mice after 8 wk on NaI
water, and from young NOD.H-2h4 mice not given NaI water, and the cDNA
was assessed for expression of cytokines and activation markers by
RT-PCR (Table V). Surprisingly,
expression of cytokine and activation marker mRNA was not reduced in
spleens of anti-TGF-treated mice compared with controls. As
shown previously (9), spleens of normal NOD.H-2h4 mice
expressed barely detectable levels of IL-2R, CD40L, IL-2, IL-4, and
IFN-, but they did express readily detectable TGF1 mRNA. These
results suggest that anti-TGF did not inhibit activation of
autoreactive T cells, although our experiments do not rule out the
possibility that T cell activation could be delayed in
anti-TGF-treated mice.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 in NOD.H-2h4
mice during development of SAT markedly inhibits development of thyroid
lesions. These results were unexpected, because TGF suppressed
several other autoimmune diseases when administered in vivo either as
the purified recombinant cytokine or by somatic gene therapy (13, 17, 18, 19, 22), and TGF produced by regulatory T cells
suppressed development of autoimmune thyroiditis in rats
(20). However, TGF1 is overproduced in
MRL/lpr mice that spontaneously develop autoimmune disease
(25, 26, 27), and lymphoid cells from some SLE patients
spontaneously produced high levels of TGF1, particularly during
disease exacerbation (26). In addition, local
administration of anti-TGF mAb suppressed synovial inflammation
induced by streptococcal cell walls in a rat arthritis model
(30), and TGF1 production in rheumatoid synovium was
important for inducing expression of the CXCR4 chemokine receptor
that promoted retention of inflammatory cells in the synovium
(31). TGF can be chemotactic for monocytes and other
inflammatory cells (14, 39, 40). Thus, local production of
TGF1, e.g., in the thyroid, could facilitate inflammation and tissue
damage by promoting the influx of inflammatory cells to a particular
tissue site and/or promoting their retention in the tissue by inducing
up-regulation of particular chemokine receptors (31).
Studies to address this possibility are in progress.
TGF can be secreted by several cell types, including T cells and
activated B cells (14, 26, 41). In MRL/lpr mice
that overproduce active TGF1, TGF1 was bound to IgG and was
secreted primarily by B cells and plasma cells (26). Serum
IgG of MRL/lpr mice was shown to bind TGF1, and the
IgG-bound TGF1 suppressed cytotoxic T cell responses to unrelated
Ags (42, 43). It was suggested that TGF1 bound to IgG
might localize and be locally activated in tissues where autoantibodies
combine with autoantigens, and thereby promote local inflammation
(42). In our experiments, TGF1 produced locally in the
thyroid and/or TGF1 produced in peripheral lymphoid organs (e.g.,
spleen) might play a role in recruiting inflammatory cells to the
thyroid. Although it is not known whether the TGF1 expressed in
thyroids of NOD.H-2h4 mice with SAT is bound to IgG, much of the
TGF1 was localized to areas of the thyroid shown previously
(9) to be comprised of clusters of
CD4+ T and B cells (Fig. 1, K and
L). The TGF1 expressed in thyroids of mice with SAT is
presumably in the active form, because the Ab used for staining has
been shown to be specific for the active form of TGF (26, 35, 36, 37). Whether the TGF is released locally in its active
form as shown by others (26) or is activated after its
release is not known.
Thyroids and spleens of normal NOD.H-2h4 mice and normal NOD.LtJ mice
constitutively express TGF1 mRNA, whereas normal thyroids and
spleens of other strains of mice constitutively express very little
TGF1 mRNA (Ref. 4 and our unpublished observations).
Others have reported that thymocytes and spleen cells of NOD mice
express 5- to 10-fold higher amounts of TGF1 mRNA compared with
C57BL/6 mice (44). This suggests that constitutive
expression of high levels of TGF1 mRNA may be characteristic of
tissues of certain strains of mice that develop SAT and/or other
autoimmune diseases. Because thyroids of normal NOD.H-2h4 or NOD.LtJ
mice have essentially no infiltrating inflammatory cells, the primary
source of TGF1 protein in SAT thyroids (Fig. 1), it is of interest
that thyrocytes of normal NOD.H-2h4 mice express readily detectable
latent TGF (Fig. 1M), whereas thyrocytes of normal CBA/J
mice do not (K. Chen, unpublished observations). Although further
studies using in situ hybridization will be needed to determine whether
normal NOD.H-2h4 thyrocytes express TGF1 mRNA, it is possible that
thyrocytes expressing high amounts of latent TGF1 are the cells in
normal thyroids that express TGF1 mRNA.
Our results suggest that TGF1, possibly produced locally in the
thyroid, has an important role in promoting thyroid inflammation in
SAT, because administration of anti-TGF results in suppression
of SAT and marked reduction of active TGF1 protein in the thyroid.
TGF1 protein is also produced in the spleens of NOD.H-2h4 mice with
SAT but is barely detectable in spleens of mice that have not developed
SAT, or in spleens of mice given anti-TGF (Table III). This
suggests that production of TGF1 in the peripheral lymphoid organs
might also be important for SAT development. Spleens of NOD.H-2h4 mice
with SAT contain activated T cells that express mRNA for activation
markers and cytokines (Ref. 9 and Table V). Because
spleens of anti-TGF-treated mice expressed activation markers
and cytokines (Table V), neutralization of TGF1 apparently did not
inhibit SAT by inhibiting the activation of autoreactive T cells. This
is consistent with the finding that anti-TGF suppressed SAT
development most effectively when it was administered weekly beginning
4 wk after the mice began NaI water. At this time, T cell activation as
evidenced by up-regulation of cytokines and activation markers in the
spleen has already occurred (9) and cells have begun to
migrate to the thyroid (4). Infiltration of inflammatory
cells in the thyroid increases considerably between 4 and 8 wk
(4). Because it is during this time that anti-TGF
most effectively inhibits SAT, TGF1 may function, at least in part,
to recruit inflammatory cells to the thyroid in this model.
Whether local production of TGF1 may also be important in promoting
retention of the inflammatory cells in the thyroid as shown in
rheumatoid arthritis (31) is not yet known. The fact that
administration of anti-TGF beginning when thyroid inflammation
was maximal did not lead to a decrease in SAT severity 3 wk later
(Table II) could suggest that TGF1 expression in the thyroid may not
be needed for retention of inflammatory cells. However, because TGF1
expression in the thyroid was very high when SAT was maximal, the
amount of Ab administered was insufficient to neutralize all the
TGF1 protein (Fig. 1, F vs G). Additional
studies, currently in progress, will be required to determine whether
increasing the amount of anti-TGF administered at the peak of
disease to completely inhibit TGF1 protein expression in the thyroid
will result in inhibition of chronic inflammation.
The requirement for TGF1 for development of SAT as
demonstrated in this study is not observed using the same mAb in an
experimentally induced mouse model of thyroiditis (EAT). In the murine
model of EAT studied in our laboratory, administration of the same
amount of anti-TGF that effectively suppressed SAT development
in NOD.H-2h4 mice had no effect on development of EAT (our unpublished
observations). This suggests that the ability of anti-TGF to
suppress thyroiditis may be unique to strains of mice that
spontaneously develop autoimmune thyroiditis and/or that constitutively
express TGF1 mRNA and relatively high levels of latent TGF1 in
tissues. It is not known whether TGF1 may also play a role in other
spontaneous models of autoimmune thyroiditis such as the obese strain
chicken. It is important to emphasize that the apparent proinflammatory
effects of TGF1 demonstrated in this work are observed with amounts
of TGF1 that are produced spontaneously during development of SAT.
It is quite possible that the supraphysiologic amounts of TGF1
administered exogenously that were shown to suppress some other
autoimmune diseases (13, 17, 18, 19, 22) would also suppress
development of SAT in this model. Further studies, currently in
progress, will be required to address this important question.
|
Acknowledgments |
|---|
1. |
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Helen Braley-Mullen, Departments of Internal Medicine and Medical Microbiology and Immunology, University of Missouri School of Medicine, M450 Medical Sciences, Columbia, MO 65212. E-mail address: mullenh{at}health.missouri.edu
3 Abbreviations used in this paper: NOD, nonobese diabetic; SAT, spontaneous autoimmune thyroiditis; MTg, mouse thyroglobulin; EAT, experimental autoimmune thyroiditis; HPRT, hypoxanthine phosphoribosyl transferase; LAP, latency-associated peptide.
Received for publication January 16, 2001. Accepted for publication October 11, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain expression in the autoimmune thyroiditis of nonobese diabetic mice. J. Immunol. 151:1691.[Abstract]
-1 (TGF1) in autoimmune diseases. J. Autoimmun. 14:23.[Medline]
. Annu. Rev. Immunol. 16:137.[Medline]
: the good, the bad, the ugly. J. Exp. Med. 180:1587.1 on immune responses of naïve versus memory CD4+ Th1/Th2 T cells. Eur. J. Immunol. 30:2101.[Medline]
1. J. Immunol. 146:3012.[Abstract]
1. J. Immunol. 146:1163.[Abstract]
suppresses acute and chronic arthritis in experimental animals. J. Clin. Invest. 87:1108.
and interleukin 4 in the prevention of autoimmune thyroiditis in rats by peripheral CD4+CD45RC- cells and CD4+CD8- thymocytes. J. Exp. Med. 189:279.-1 enhances the clinical severity of disease. J. Neuroimmunol. 47:1.[Medline]
1 somatic gene therapy prevents autoimmune disease in nonobese diabetic mice. J. Immunol. 161:3950. enhances the in vivo effector function and memory phenotype of antigen-specific Th cells in experimental autoimmune encephalomyelitis. J. Immunol. 148:2109.[Abstract]
has contrasting effects in the
presence or absence of exogenous IL-12 on the in vitro activation of
cells that transfer experimental autoimmune thyroiditis. J.
Interferon Cytokine Res. 21:971.
suppresses host defense against bacterial infection in MRL/lpr mice. J. Exp. Med. 180:1693.1 in B cells and plasma cells of autoimmune mice: IgG bound TGF-1 suppresses neutrophil function and host defense against Staphylococcus aureus infection. J. Clin. Invest. 98:2496.[Medline]
in neonatal and adult MRL-lpr mice. J. Autoimmun. 9:463.[Medline]
1 overexpression in murine pancreas induces chronic pancreatitis and, together with TNF-
, triggers insulin-dependent diabetes. Biochem. Biophys. Res. Commun. 26:13279.
. J. Exp. Med. 171:231.. J. Exp. Med. 177:225.1 on synovial T cells contributes to their accumulation within the rheumatoid synovium. J. Immunol. 165:3423.: bioactivity, neutralization, and TGF-2 affinity purification. J. Immunol. 142:1536.[Abstract]
in situ using engineered tissue. Am. J. Pathol. 147:1228.[Abstract]
1
activation in situ: quantitative and functional evidence after low-dose
gamma-irradiation. FASEB J. 11:991.
1 is required for secretion of IgG of all subclasses by LPS-activated murine B cells in vitro. J. Immunol. 151:4625.[Abstract]
) induces monocyte chemotaxis and growth factor production. Proc. Natl. Acad. Sci. 84:5788. in the pathophysiology of chronic inflammation. J. Periodontol. 64:450.[Medline]
. FASEB J. 12:308. (Abstr.).
and suppress cytolytic T cell responses. J. Exp. Med. 181:1875.-like immunosuppressive factor in immunoglobulin G-binding factor. J. Exp. Med. 182:1717.This article has been cited by other articles:
|
|
 |
|
  V. Saxena, D. W. Lienesch, M. Zhou, R. Bommireddy, M. Azhar, T. Doetschman, and R. R. Singh Dual Roles of Immunoregulatory Cytokine TGF-{beta} in the Pathogenesis of Autoimmunity-Mediated Organ Damage J. Immunol., February 1, 2008; 180(3): 1903 - 1912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. McLachlan, H. Braley-Mullen, C.-R. Chen, H. Aliesky, P. N. Pichurin, and B. Rapoport Dissociation between Iodide-Induced Thyroiditis and Antibody-Mediated Hyperthyroidism in NOD.H-2h4 Mice Endocrinology, January 1, 2005; 146(1): 294 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Olivieri, S. De Angelis, V. Vaccari, H. Valensise, F. Magnani, M. A. Stazi, R. Cotichini, E. Gilardi, V. Cordeddu, M. Sorcini, et al. Postpartum Thyroiditis Is Associated with Fluctuations in Transforming Growth Factor-{beta}1 Serum Levels J. Clin. Endocrinol. Metab., March 1, 2003; 88(3): 1280 - 1284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Chen, Y. Wei, G. C. Sharp, and H. Braley-Mullen Inhibition of TGF{beta}1 by Anti-TGF{beta}1 Antibody or Lisinopril Reduces Thyroid Fibrosis in Granulomatous Experimental Autoimmune Thyroiditis J. Immunol., December 1, 2002; 169(11): 6530 - 6538. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||
