| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
B
Cell Regulation Section (S.-I.T., M.S., C.G., G.N., M.S., V.M., K.S., L.D.K.) Metabolic Diseases Branch National Institute of Diabetes and Digestive and Kidney Diseases and Experimental Immunology Branch (D.S.S.) National Cancer Institute National Institutes of Health Bethesda, Maryland 20892
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
B and a c-fos family member, fra-2, which was
previously shown to be important in the suppression of class I levels
by hydrocortisone. Unlike hydrocortisone, iodide also increases the
formation of a complex with enhancer A, which we show, in antibody
shift experiments, is a heterodimer of the p50 and p65 subunits of
NF-B. The changes in these complexes are not duplicated by chloride
and are related to the action of iodide on class I RNA levels by the
following observations. First, FRTL-5 thyroid cells with an aged
phenotype coincidentally lose the ability of iodide to decrease MHC
class I RNA levels and to induce changes in either complex. Second, the
effect of iodide on class I RNA levels and on enhancer A complex
formation with Mod-1 and the p50/p65 heterodimer is inhibited by agents
that block the inositol phosphate, Ca++,
phospholipase A2, arachidonate signal transduction pathway:
acetylsalicylate, indomethacin, and 5,8,11,14-eicosatetraynoic acid.
Interestingly, iodide can also decrease formation of the Mod-1 complex
and increase formation of the complex with the p50/p65 subunits of
NF-B when the NF-B enhancer sequence from the Ig light chain,
rather than enhancer A, is used as probe; and both actions mimic the
action of a phorbol ester. This suggests that iodide may regulate
complex formation with NF-B regulatory elements on multiple genes
associated with growth and function, providing a potential mechanism
relating the autoregulatory action of iodide on thyroid cells and its
action on class I gene expression. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-Interferon,
which increases class I but does not cause aberrant class II expression
(6), can mimic autoimmune changes in class I levels in thyrocytes and
other cells in culture (1, 2, 7) and can enhance or induce the
appearance of thyroid autoantibodies and ATD when used to treat
patients with hepatitis or cancer (8, 9, 10). Most importantly, class I
deficient mice do not develop autoimmunity in several disease models,
i.e. systemic lupus erythematosus (11), diabetes (12, 13),
or autoimmune blepharitis (14); and methimazole, a drug used to treat
patients with Graves’ disease, can suppress class I levels in rat
FRTL-5 thyroid cells (7) and mimic the class I deficient state to
prevent systemic lupus erythematosus and autoimmune blepharitis in
those experimental models (14, 15). High concentrations of iodide have been used in the past to treat patients with Graves’ disease; although only transiently effective, iodide is still used to prepare patients for surgery (16). The therapeutic action of high concentrations of iodide in ATD had been linked to its autoregulatory action on thyroid growth and function both in vivo (17) and in cultured FRTL-5 cells (18, 19, 20). In recent reports, we noted that high concentrations of iodide could also suppress class I levels in rat FRTL-5 and human thyroid cells (7, 21) and that as little as 4 days of high-iodide therapy could suppress class I levels in the thyroids of Graves’ patients being prepared for surgery (21). Coupled with the accumulating evidence for a role in class I overexpression in ATD, this suggested that the ability of iodide to suppress class I might be important in its therapeutic action in patients with Graves’ disease and that rat FRTL-5 cells might be a useful model with which to study this phenomenon.
FRTL-5 thyroid cells are a continuously cultured line whose growth and function, like other thyroid cells, depend not only on TSH/cAMP, but also on hydrocortisone, insulin, insulin-like growth factor-I (IGF-I), and other factors in serum (22, 23, 24). We recently showed that MHC class I expression in FRTL-5 cells was decreased by TSH/cAMP, hydrocortisone, insulin, IGF-I, and/or serum (7, 25 25A ), i.e. by all the hormones and factors necessary for thyroid cell growth and function. We hypothesized 1) that hormonal suppression of class I levels might be a normal mechanism to preserve self-tolerance in the face of hormone-induced increases in genes important for growth and function and 2) that common transcription factors might be involved in these coordinate actions (3, 7, 25 25A ). Understanding the mechanism of iodide suppression of class I gene expression in FRTL-5 cells might, therefore, enhance our understanding of the transcriptional mechanisms underlying self-tolerance and the autoregulatory actions of iodide on growth and function.
This report identifies a transcriptional mechanism by which iodide
suppresses MHC class I gene expression in thyrocytes; it is the first
description of a transcriptional action of iodide. We show that iodide
regulates interactions between the ubiquitous transcription factor,
NF-B, and enhancer A of the MHC class I gene; and we link this
action to the phosphoinositide-Ca++-arachidonate signal
pathway, which is known to be involved in the growth and function of
FRTL-5 thyroid cells (22). We show that the action of iodide is not
restricted to enhancer A of the class I gene, but can similarly
modulate complex formation with NF-B elements on many genes
implicated in the regulation of cell growth and function by ligands
such as phorbol esters. A preliminary report of these data has been
presented (25a).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). This rapid action is consistent with
the rapid effects of high concentrations of iodide in clinical practice
(17) and on growth or DNA synthesis in FRTL-5 cells (18, 19, 20). The
effect of a single addition of iodide begins to wane after 24 h
and is nonexistent 96 h after challenge (Fig. 1B); however, the
decrease will persist over a 96-h period if fresh iodide is added at
24, 48, and 72 h after the start of the experiment (Table 1). Interestingly, the effect of iodide
is lost in FRTL-5 thyroid cells with an aged phenotype, i.e.
in cells with passage numbers > 30 (Fig. 1, A and B).
|
|
). This may be relevant to the
fact that autoimmune thyroid disease tends to occur with aging, even in
families with a history of thyroid autoimmunity.
The ability of iodide to suppress class I levels was dependent on the
concentration of iodide and was evident at higher iodide concentrations
associated with iodide autoregulation in these cells (17, 18, 19, 20);
inhibition was only 21 ± 6% of control values at 0.1
mM iodide, but 51 ± 8% at 0.5 mM,
66 ± 4% at 1 mM, and 78–80 ± 7% at 5 or 10
mM. Data are the mean ± SE from four
independent experiments and represent a significant decrease in RNA
(P < 0.05 or better). Suppression was not duplicated
by chloride ions, which are present at >100 mM
concentrations in the controls without iodide. The effect was evident
in FRTL-5 cells maintained with (data not shown) or without (Fig. 1)
TSH in the medium.
The Effect of Iodide on Class I RNA Levels Reflects an Action
Involving Enhancer A of the Class-I Promoter
Iodide Regulates Promoter Activity via Enhancer A
At a concentration of 1 mM, which is effective in
decreasing endogenous class I RNA levels (Fig. 1), iodide decreased the
activity of an exogenous class I promoter containing 1100 bp of class I
5'-flanking region that had been ligated to a chloramphenicol
acetyltransferase (CAT) reporter gene, p(-1100)CAT, and that had been
transfected into FRTL-5 thyroid cells (Fig. 2). Using 5'-deletion mutants, the iodide
effect could be localized to between -203 and -127 bp of the start of
transcription (Fig. 2). This region contains two major regulatory
elements, enhancer A, -180 to -170 bp, and the interferon response
element, -161 to -150 bp (Fig. 2, bottom).
|
) and
was lost in a construct, p(-203 M)CAT, which has a
mutation (termed Mutant 2) of enhancer A, from 5'-GGGGAGTCCCC-3' (-180
to -170 bp) to 5'-GCCGAGTCAAG-3' (Fig. 2 and
Table 2). The same result, a loss of
iodide suppression, was also seen using another p(-203)CAT enhancer A
mutation (termed Mutant 1) to 5'-GCGGAGTCAAG-3'
(Table 2). Enhancer A is the element we linked to the ability of
hydrocortisone to suppress MHC class I gene expression; the same
mutations eliminate the ability of hydrocortisone to decrease MHC class
I gene expression (28).
|
).
This suggested that the iodide effect on class I gene expression, like
hydrocortisone, involved the enhancer A element.
Iodide Regulates the Formation of Two Protein/DNA Complexes with
Enhancer A: Mod-1 and a p50/p65 Heterodimer
Hydrocortisone decreases formation of a protein complex with enhancer
A, -180 to -170 bp, in the class I 5'-flanking region, termed Mod-1,
which involves the p50 subunit of NF-B and a c-fos family
member, fra-2 (28). Using the 74-bp region between -203 and
-130 bp as a labeled probe, we could show that iodide treatment of
FRTL-5 cells for as little as 1.5 h resulted in extracts with a
reduced ability to generate a protein/DNA complex located toward the
top of the electrophoretic mobility shift analysis (EMSA) gel (Fig. 3, lane 3 vs. 2). This complex
was identified as Mod-1 by the ability of antibodies to the p50 subunit
of NF-B (Fig. 3, lane 4) and to fra-2 (Fig. 3, lane 9) to
inhibit its formation and/or supershift the complex decreased by
iodide. Antibodies to the p65 subunit of NF-B, p52,
c-fos, fra-1, or c-jun had no effect
on the Mod-1 complex (Fig. 3, lanes 5 to 8 and 10, respectively).
|
). In this experiment,
iodide treatment of FRTL-5 cells decreased formation of a major complex
formed with these extracts (Fig. 4A, lane 2 vs. 1). This is
the Mod-1 complex, as evidenced by the ability of antibodies to the p50
subunit of NF-B (Fig. 4B, lane 3) or fra-2 (Fig. 4B, lane
7) to prevent its formation and/or supershift the complex, but not
antibodies to the p65 subunit of NF-B, p52, c-fos,
fra-1, or c-jun (Fig. 4B, lanes 4 to 6 and 8 and
9, respectively). These studies showed the effect of the two antibodies
on the complexes in extracts from cells before iodide treatment,
whereas the data in Fig. 3 showed the effect of the antibodies on the
complexes in cells treated with iodide. The data thus indicate that the
antibody specificities are not affected by iodide treatment of the
cells. More importantly, the data in both figures show that there is no
complex in this region that is not Mod-1 and that the Mod-1 complex
with enhancer A is decreased by iodide treatment of the cells.
Consistent with this, formation of the complex was inhibited by the
presence of a 100-fold excess of unlabeled enhancer A oligonucleotide
in the binding mixture (Fig. 4B, lane 11 vs. 10), but not by
a 100-fold excess of oligonucleotides with the mutations in the
enhancer A site (Fig. 4B, lanes 12 and 13) that result in a loss of the
ability of iodide to suppress class I promoter activity (Fig. 4 and
Table 2).
|
B. This is measurable
using conditions, detergents and high salts, that favor formation of
this complex, but appear to restrict formation of the Mod-1 complex.
Thus, iodide increases formation of a complex with the radiolabeled
enhancer A oligonucleotide (Fig. 5A, lane
3 vs. 2) in the presence of detergent and high salts. The
complex is specific, since its formation is prevented by unlabeled
enhancer A (Fig. 5A, lane 4) but not by either of its mutants (Fig. 5A, lanes 5 and 6), whose sequences are noted on the bottom of
Fig. 4 and which cause a loss of iodide-induced suppression of class I
promoter activity (Fig. 2 and Table 2). The complex is decreased and
supershifted by an antibody to the p50 subunit of NF-B and decreased
by an antibody to the p65 subunit of NF-B (Fig. 5A, lanes 9 and 10,
respectively), but unaffected by antibodies to c-rel,
fra-1, fra-2, c-fos, or
c-jun (Fig. 5A, lanes 11–15).
|
and 4) or increase formation of the p50/p65
heterodimer complex (Fig. 5A) was rapid, preceding the change in RNA
(Fig. 1), and was consistent with the changes in class I RNA with time
after a single exposure to iodide (Fig. 1). Thus it was evident within
60 min of iodide (1 mM) challenge, was maximal at 2 h,
decreased by 3–6 h, and, although still evident at 24 h, as
illustrated for p50/p65 heterodimer formation (Fig. 5B, lanes 3–10)
after a single exposure to iodide, was beginning to wane. This is very
different from the action of hydrocortisone in two respects.
Hydrocortisone does not increase p50/p65 complex formation with
enhancer A when evaluated under comparable conditions; and the ability
of hydrocortisone to decrease Mod-1 requires 24–48 h of FRTL-5 cell
treatment, comparable to its effect as a function of time on RNA levels
and promoter activity (28). The ability of iodide to both increase
p50/p65 complex formation (Fig. 5A) or decrease Mod-1 (Figs. 3 and 4)
was also evident only at higher iodide concentrations, 0.1 to 10
mM, like the class I RNA decrease, as illustrated for
p50/p65 heterodimer formation (Fig. 5B, lanes 12–14); it was not
evident at 0.01 mM or lower concentrations of iodide (data
not shown).
Second, the effect of iodide on formation of the p50/p65 heterodimer
complex is very specific and does not reflect a generalized action on
NF-B subunits. Thus, a second, non-iodide-induced, faster moving
complex with enhancer A is inhibited only by anti-p50 (Fig. 5A, lane
9). This property, plus its migration on gels with respect to different
concentrations of authentic p50 protein (data not shown), is consistent
with its identification as a p50 homodimer. Iodide regulates,
therefore, formation of only the p50/p65 heterodimer complex with
enhancer A and not formation of the p50 homodimer complex.
Last, enhancer A has a core sequence, GGGGA, which is
common to NF-B binding elements from other genes, e.g.
the enhancer sequence from the Ig light chain. A 22 mer
oligonucleotide containing the enhancer sequence from the Ig light
chain can prevent formation of both the iodide-induced p50/p65
heterodimer and p50 homodimer (Fig. 5A, lane 7), but not the lower
complexes. We previously showed that the same 22 mer oligonucleotide
containing the enhancer sequence from the Ig light chain could
prevent formation of Mod-1 (28). With the exception of the core
GGGGA sequence, the sequence of the oligonucleotide
containing enhancer A of class I,
5'-CGGTGGTGGGGAGTCCCCGTGTCC-3', differs from
5'-AGTTGAGGGGACTTTCCCAGGC-3', the Ig light
chain enhancer, in that it has a longer inverted repeat
(italicized) and a different spacing to the inverted repeat.
Thus, the GGGGA is a critical element in the formation of
the iodide-induced p50/p65 heterodimer complex, just as it is for the
iodide-decreased Mod-1 complex, as previously demonstrated (28). The
import of this observation will be further noted in experiments to be
described and discussed below.
The Effect of Iodide on Class I Gene Expression Is Linked to Its Effect on Complex Formation; Both Effects Are Mediated by an Action Involving the Phosphoinositide-Ca++-Arachidonic Acid-Cyclooxygenase Signal Transduction Pathway
The ability of iodide to decrease Mod-1 and increase the p50/p65 heterodimer complex was functionally linked to the action of iodide on class I RNA levels by the following experiments.
First, we had shown that iodide did not decrease class I RNA levels in
FRTL-5 cells with an aged phenotype (Fig. 1). Iodide also did not
decrease the Mod-1 complex in FRTL-5 cells with an aged phenotype (Fig. 6A, lanes 3 vs. 2 or 5
vs. 4); nor did iodide increase the formation of the p50/p65
heterodimer complex (Fig. 6B, lanes 3 vs. 2 or 5
vs. 4). This was not true for a phorbol ester,
o-tetradecanoyl phorbol 13-acetate (TPA), which can increase formation
of the p50/p65 heterodimer and the p50 homodimer (Fig. 6B, lane 6
vs. 2) in aged cells, unlike its vehicle, dimethylsulfoxide
(Fig. 6B, lane 7 vs. 2), and served as a positive control.
These data indicate that both the iodide-induced decrease in the Mod-1
complex with enhancer A and the iodide-induced increase in the p50/p65
heterodimer complex are lost concurrently with the loss in
iodide-induced decreases in class I RNA levels. The effect of the
phorbol ester on class I RNA levels is similar to that of iodide, as
preliminarily reported (28a); this will be presented separately (S.-I.
Taniquchi, M. Shong, V. Montani, C. Giuliani, M. Saji, D. S. Singer,
and L. D. Kohn, manuscript in preparation).
|
As illustrated in assays measuring p50/p65 heterodimer and p50
homodimer complex formation with enhancer A, the ability of iodide to
increase complex formation (Fig. 7A, lane
7 vs. 6; Fig. 7B, lane 5 vs. 2) in extracts from
cells with low passage numbers was inhibited by the phospholipase
A2 (PLA2) inhibitor, acetyl salicylate (ASA);
by the cyclooxygenase inhibitor, indomethacin (Fig. 7A, lanes 8 and 9,
respectively, vs. lane 7); and by 5,8,11,14-eicosatetraynoic
acid (ETYA) (Fig. 7B, lane 6 vs. 5), which inhibits
cyclooxygenase, lipoxygenase, epoxygenase, and PLA2
activities. The ability of these agents to inhibit iodide-induced
p50/p65 heterodimer complex formation appeared to be specific, since
none of the agents blocked the ability of a maximally effective
concentration of TPA, 50 nM, to increase p50/p65
heterodimer or p50 homodimer complex formation (Fig. 7A, lanes 4 and 5
vs. lane 3; Fig. 7B, lane 4 vs. 3). Additionally,
none of the agents, ASA, indomethacin, or ETYA, inhibited basal complex
formation in control extracts (Fig. 7C). Moreover, the ability of ASA,
indomethacin, and ETYA to inhibit iodide-increased p50/p65 heterodimer
complex formation with enhancer A was coincident with their abilities
to inhibit the ability of iodide to decrease class I RNA levels and
decrease Mod-1 complex formation and was again specific with respect to
the action of TPA (Table 3).
|
|
and 7 and Table 3) support the following
conclusions. The iodide-induced changes in the Mod-1 and p50/p65
complexes with enhancer A are involved in class I suppression, since
they correlate with iodide-induced changes in RNA levels. Iodide acts
to alter complex formation and decrease class I levels by activating
the phospholipase A2-arachidonic acid signal transduction
system (Fig. 7D). Additionally, iodide modulation of complex formation
differs mechanistically from that of a phorbol ester, since TPA
activity is not inhibited by ASA, indomethacin, or ETYA (Fig. 7D).
Phorbol esters appear to act via protein kinase C; as noted above, a
separate report will address the effect of phorbol esters on class I
activity and has been presented in preliminary fashion (28a).
Two points should be noted before continuing. The persistent
ability of TPA to alter complex formation in FRTL-5 cells with an aged
phenotype (Fig. 6), which have an altered inositol
phosphate-Ca++-arachidonate-cyclooxygenase signal pathway
(26), is consistent with the insensitivity of TPA action to inhibitors
of this pathway (Fig. 7). Additionally, and also consistent with the
data in Figs. 6 and 7, the PLA2 activator, mellitin (5
µg/ml), as well as the arachidonic acid metabolites prostaglandin
E2 (10 µM) and prostaglandin
F2 (10 µM), increased p50/p65 heterodimer
complex formation in FRTL-5 cells with a low passage number but not in
cells with a high passage number (data not shown).
Iodide Regulates Complex Formation with Elements from Other Genes
Interacting with the Subunits of NF-B
Enhancer A of the class I promoter has a core sequence that
is present in NF-B binding elements from other genes,
e.g. the enhancer sequence from the Ig light chain (Fig. 8, top). As noted above (Fig. 5A), an oligonucleotide containing the core sequence of the NF-B
binding element of the enhancer sequence from the Ig light chain
will prevent formation of the iodide-induced p50/p65 complex with the
enhancer A oligonucleotide from the class I gene. We therefore
evaluated the effect of iodide on complex formation with the NF-B
binding element to determine whether iodide might, as a first
approximation, be capable of regulating genes other than MHC class I in
thyrocytes.
|
B binding element from the Ig
light chain will form a protein/DNA complex at the top of the gel under
low salt conditions without detergent, which is decreased in extracts
from iodide-treated cells (Fig. 8A, lane 3 vs. 2). This is
Mod-1 as evidenced by the ability of anti-p50 and anti-fra-2
to inhibit its formation and supershift the complex (Fig. 8A, lanes 5
and 6, respectively), but not antibodies to p65, fra-1,
c-fos, or c-jun (data not shown).
Additionally, extracts from iodide-treated cells will form a
p50/p65 heterodimer complex with the NF-B binding element when
incubated in high salts plus detergent (Fig. 8B, lane 3 vs.
2). The identification of the p50/p65 complex, whose formation is
induced by iodide, is evidenced by its reactivity with anti-p50 (Fig. 8B, lane 4) and anti-p65 (Fig. 8B, lane 5) but not anti-p52,
anti-c-rel, or anti-fra-2 (Fig. 8B, lanes 6–8).
The effect of iodide treatment of the cells is rapid, as evidenced by
induction of the p50/p65 heterodimer complex (Fig. 8C), and depends on
high iodide concentrations (Fig. 8D) as is the case for the formation
of this complex with enhancer A (Fig. 5, A and B), although slight
differences cannot be excluded.
The effect of iodide on both Mod-1 and the p50/p65 heterodimer
complex formed by the NF-B binding element from the Ig light
chain is duplicated by a phorbol ester (see
Figs. 3–7). Further, and
again unlike iodide, the phorbol ester increases p50 homodimer
formation as illustrated in Fig. 7, A and B. These complexes were again
identified using the antibodies.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Iodide therapy in Graves’ disease is a well recognized treatment whose
effectiveness is associated with the Wolf-Chaikoff effect (17, 36, 37),
wherein, for example, high concentrations of iodide suppress thyroid
hormone secretion. Iodide therapy is additionally associated with the
autoregulatory action of iodide that affects the growth of the thyroid
(16, 17, 18, 19, 20). The mechanisms underlying both the iodide-immunosuppressive
action in Graves’ patients and iodide autoregulation of growth and
function remain largely unclear (16, 17, 37). In recent studies (7, 21), we showed that iodide at high concentrations, the same as those
used to study its autoregulatory action (18, 19, 20, 38), could decrease
MHC class I surface expression and/or RNA levels in human and rat
FRTL-5 thyrocytes. Moreover, we showed that preoperative iodide
treatment of Graves’ patients suppressed MHC class I RNA levels in the
Graves’ thyroid (21). In this report we describe a transcriptional
action of iodide that can account for its immunosuppressive action on
MHC class I gene expression in Graves’ and that provides a potential
explanation for its complex autoregulatory action on growth and
function by its relationship to the transcriptional mechanism of action
of phorbol esters. Thus, this is the first recognition that iodide
regulation of class I gene expression and thyroid growth or function,
like phorbol esters, involves the NF-B system and the complex array
of its subunit interactions with the jun/fos or
rel transcription factors.
We show in this report that high concentrations of iodide, like
hydrocortisone (28), can decrease the formation of a protein/DNA
complex, termed Mod-1, with enhancer A of the class I promoter. Unlike
the case of hydrocortisone (28), however, we show that iodide can
simultaneously increase the formation of a new and different complex
with enhancer A, which, like Mod-1, involves the subunits of the
NF-B-regulatory transcription factor. Mod-1 is a complex (28)
involving the p50 subunit of NF-B and a member of the a
fos family member, fra-2 (39); the new complex is
a heterodimer of the p50 and p65 subunits of NF-B. We show that a
critical recognition element for the binding of both complexes is the
GGGGA sequence within enhancer A and that the iodide action
on the formation of each complex and on enhancer A-dependent class I
promoter activity correlates with its ability to suppress MHC class I
RNA levels. Thus, we show that aged cells concurrently lose these
activities and that inhibitors of the ability of iodide to modulate the
arachidonate signal transduction system also concurrently inhibit
them.
Hydrocortisone suppresses class I by decreasing the formation of the Mod-1 complex with enhancer A, -180 to -170 bp in its 5'-flanking region (28). In contrast, interferon increases the binding of Mod-1 to enhancer A, in association with its ability to increase class I expression (28). Enhancer A is, therefore, an important element involved in the interferon response, in addition to the interferon response element (28, 40, 41); suppression of the Mod-1 interaction with enhancer A by anti-immune agents, such as hydrocortisone, is one means of counteracting the action of interferon and, by extension, the autoimmune response. The basis for this appears to be the fact that the p50 component of Mod-1 has a footprint that overlaps the enhancer A and interferon response elements (28). These data are consistent with separate data which showed that down-regulation of the p50 subunit decreases MHC class I expression (42).
We suggest, therefore, that iodide, like hydrocortisone (28), decreases
the Mod-1 complex with enhancer A, thereby decreasing MHC class I
promoter activity and counterbalancing the action of cytokines such as
interferon that increase complex formation and class I expression (Fig. 9). Unlike hydrocortisone, iodide also
increases the formation of a protein/DNA complex between enhancer A and
a p50/p65 heterodimer of NF-B subunits (Fig. 9), and this action, as
well as a rapid decrease in Mod-1, is duplicated by phorbol esters,
which also can suppress class I activity and RNA levels in FRTL-5
thyroid cells, as preliminarily reported (28a).
|
)? Does it play
a more important role in the activation of other genes with the ability
to bind the p50/p65 heterodimer of NF-B, as is the case for phorbol
esters?
NF-B is an inducible eukaryotic transcription factor of the
rel family. It exists in the cytoplasm of most cells in an
inactive form, where it is bound to IB (43, 44, 45). NF-B is
activated in response to a number of stimulants including phorbol
esters, tumor necrosis factor- (TNF-), and interleukin-1
(43, 44, 45). Stimulation triggers the release of NF-B from IB,
resulting in the translocation of its subunits, p50 and p65, into the
nucleus. The subunits of NF-B are now known to form homodimers or
heterodimers, interact with subunits from the fos and
jun protooncogene families, and interact with other members
of the rel family to regulate the transcription of multiple
genes involved in immune and growth responses of cells (42, 43, 44, 45). Thus,
they are implicated in cytokine activation (interleukins-1, -6, and -8,
and interferon-ß, for example), cell adhesion (regulation of
endothelial-leukocyte adhesion molecule-1, intercellular adhesion
molecule-1, and vascular cell adhesion molecule-1), and T cell growth,
as well as commitment. In short, they are implicated in the expression
of a pleiotropic array of genes induced in response to such agents as
growth factors, mitogens, tumor promoters, cytokines, TNF-,
and antioxidants, i.e. genes that are linked to both
immunomodulation as well as growth regulation. It is, therefore, not
surprising, in retrospect, that high concentrations of iodide would
have both immunomodulatory actions and autoregulatory effects on the
growth and function of the thyroid cell.
In studies of lymphocytes, sodium salicylate has been reported to
inhibit the activation of NF-B by interfering with the pathway that
leads to the phosphorylation or degradation (or both) of IB,
i.e. its action is suggested to be at the level of the
release of NF-B from IB and is opposite to the action of a
phorbol ester such as TPA (46). In those studies (46), the sodium
salicylate action was not duplicated by indomethacin or ETYA; hence the
involvement of the phosphoinositide-Ca++-arachidonate
system was excluded. This is not the case for iodide in the present
study in thyrocytes. Thus, iodide would appear to have a direct
transcriptional effect resulting from its activation of the
phosphoinositide-Ca++-arachidonate pathway in thyroid
cells; and the action of salicylates may be more complex than perceived
in the earlier pioneering experiments (46). These data may reflect the
importance of this pathway to the growth and function of thyrocytes
(22, 23, 26, 27); additionally, the complexity of changes in p50 and
p65 complexes noted in this report, i.e. decreases in Mod-1,
increases in the p50/p65 heterodimer, but no change in the p50
homodimer, would not likely be caused simply by a change in NF-B
release from IB and a mass action effect. A similar case can be made
for the effect of the phorbol esters, i.e. the possibility
must be considered that the action of TPA is more diverse than simply
mediating the release of NF-B from IB.
The action of iodide to suppress class I and ATD, autoregulate thyroid
growth, and act via NF-B, with its pleiotropic effects on genes
implicated in both the immune and growth response of the cell, also
raises the specter of tissue-specific control. Thus, iodide appears to
represent a thyroid-specific action on a ubiquitous regulatory system,
since its effect is lost in the FRT thyroid cell, whose thyroid
function is lost coincident with the loss of a thyroid-restricted
transcription factor, thyroid transcription factor-1 (7). This cannot
be related solely to the loss of the TSH-induced iodide symporter,
since iodide action on class I is still evident in 5H medium in which
TSH is not present. Continued study of the action of iodide on the
NF-B system of the thyroid may, therefore, contribute to our
understanding of the tissue-specific controls on this ubiquitous
regulatory system.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-32P]deoxy-CTP (3000 Ci/mmol),
[14C]chloramphenicol (50 mCi/mmol), and
[
-32P]ATP were from Amersham (Arlington Heights, IL).
The p50 subunit of NF-B was from Promega (Madison, WI); antibodies
to the p50 and p65 subunits of NF-B, c-rel, the
c-fos family, and c-jun were from Santa Cruz
Biotechnology, Inc. (Santa Cruz, CA). The NF-B consensus binding
site, 5'-AGTTGAGGGGACTTTCCCAGGC-3', was from Promega; the 24 mer,
5'-CGGTGGTGGGGAGTCCCCGTGTCC-3', encompassing enhancer A, as well as its
mutants, was from Operon Technologies, Inc. (Alameda, CA).
Cell Culture
FRTL-5 rat thyroid cells (Interthyr Research
Foundation, Baltimore, MD; ATCC No. CRL 8305) were a fresh subclone
(F1) that had all properties previously detailed (7, 21, 24, 25 25A, 28, 47, 48, 49). Thus, their doubling time with TSH was 36 ± 6 h;
and without TSH, they did not proliferate. After 6 days in medium with
no TSH, addition of 1 x 10-10 M TSH
resulted in 10-fold or better increases in thymidine incorporation into
DNA. Fresh cells were diploid and between their fifth and 20th passage.
FRTL-5 thyroid cells with an aged phenotype had been passaged 30 or
more times (26). Like fresh phenotype cells, they were diploid and did
not proliferate without TSH. However, in the absence of TSH, their
basal thymidine incorporation into DNA during 72 h was increased
at least 10-fold over fresh phenotype cells and, after 6 days in medium
with no TSH, addition of 1 x 10-10 M TSH
resulted in only an approximately 2-fold increase in thymidine
incorporation into DNA. Their doubling time with TSH was about 26
± 6 h.
All cells were grown in 6H medium consisting of Coon’s modified F12 (Sigma Chemical Co.) supplemented with 5% calf serum, 1 mM nonessential amino acids (GIBCO, Grand Island, NY), and a mixture of six hormones: bovine TSH (1 x 10-10 M), insulin (10 µg/ml), cortisol (0.4 ng/ml), transferrin (5 µg/ml), glycyl-L-histidyl-L-lysine acetate (10 ng/ml), and somatostatin (10 ng/ml) (48, 49). Fresh medium was added every 2 or 3 days, and cells were passaged every 7–10 days. In different experiments, as noted, cells were maintained in 5H medium, which contains no TSH.
RNA Isolation and Northern Analysis
Total cellular RNA was isolated, Northern analyses were
performed, and filters were sequentially hybridized with the following
cDNA probes (0.5–1.0 x 106 cpm/ml) (7, 25 25A, 28, 50):
a 1.0-kb HpaI fragment of the MHC class I clone spanning the
entire cDNA insert (7, 25 25A, 28) and ß-actin, which was provided by Dr.
B. Paterson, National Cancer Institute, NIH.
Construction of MHC Class I Promoter-CAT Chimeric Plasmids
The CAT chimeras of the PD1 swine 5'-flanking sequences,
p(-1100)CAT, p(-549)CAT, p(-203)CAT, and p(-127)CAT have been
described (28, 47); they are numbered from the nucleotide at the 5'-end
to +1 bp, the start of transcription (28, 47). CAT constructs with
mutated enhancer A sequences were created by two-step, recombinant PCR
methods (28, 51, 52). In the first step, two PCR products that overlap
the sequence were created, both of which contain the same mutation
introduced as part of the PCR primers. The second step PCR was
performed using these overlapped PCR products as template and DNA
sequence of the 5'- or 3'-end of the final products as primer. The PCR
products were inserted into the multicloning site of pSV3CAT (28). A
Perkin-Elmer Cetus (Norwalk, CT) thermal cycler was used; the reaction
was performed at 94 C (1 min), 55 C (2 min), and 72 C (3 min) for 30
cycles; final extension was for 7 min at 72 C. The amplified fragment
was purified using 1.5% agarose gel electrophoresis.
The sequences of all constructs were confirmed by a standard method (53); DNA was prepared and twice purified by CsCl gradient centrifugation (54).
Transfection
FRTL-5 cells stably transfected with class I promoter-CAT
chimeras have been described (28). To test the effect of iodide, cells
were grown to 70–80% confluency in 6H medium and then maintained
without TSH (5H medium) for 6 days, at which time they were exposed to
1 mM iodide for 24 h before CAT activity was measured.
Transient transfections used the same class I-CAT chimeras and a
diethylaminoethyl (DEAE)-dextran procedure (28, 55). Cells were grown
to near (80%) confluency in 6H medium, shifted to 5H medium for 1 day,
washed twice with Dulbecco’s modified PBS (DPBS), pH 7.4, and
incubated 1 h with 5 ml serum-free 5H medium containing the
plasmid DNA plus 250 µg DEAE-dextran (5 Prime
3 Prime, Inc.,
Boulder, CO). Cells were then exposed to 10% dimethylsulfoxide in DPBS
for 3 min, washed twice in DPBS, cultured in 5H medium for 48 h,
and then maintained therein another 24 h with or without iodide as
noted. Efficiency of transfection was determined by cotransfection with
5 µg pRSVLuc, kindly provided by Dr. S. Subramani, University of
California (La Jolla, CA) (56). CAT values, mean ± SE
of three experiments, are normalized to luciferase activity and protein
using the Promega assay system and a Moonlight 2010 luminometer (28).
Cell viability was approximately 80% in all experiments. CAT assays
were performed as described (28, 47, 57).
Cell and Nuclear Extracts
FRTL-5 cells were grown in the presence of complete 6H medium
(+TSH) until 80% confluent and then maintained in 5H medium (-TSH)
with 5% calf serum. Cells were exposed to iodide or TPA, in the
presence or absence of ASA, indomethacin, or ETYA at the concentrations
noted. Unless otherwise noted, cellular extracts were prepared by a
modification of a described method (28, 47, 58) after scraping and
centrifuging (500 x g) cells that had been washed
twice in cold DPBS, pH 7.4. The pellet was resuspended in 2 vol of
Dignam buffer C (80) [25% glycerol, 20 mM HEPES-KOH, pH
7.9, 1.5 mM MgCl2, 0.42 M NaCl, 0.5
mM dithiothreitol (DTT), 1 µg/ml leupeptin, 1 µg/ml
pepstatin, and 0.5 mM phenyl-methylsulfonyl fluoride
(PMSF)]. The final NaCl concentration was adjusted on the basis of
cell pellet volume to 0.42 M. Cells were lysed by seven
repeated cycles of freezing and thawing. The extracts were centrifuged
at 32,500 rpm (100,000 x g) and at 4 C for 20 min. The
supernatant was recovered, aliquoted, and stored at -70 C.
Nuclear extracts were prepared as described (28, 59, 60) from identically treated and harvested FRTL-5 cells. After centrifugation at 500 x g, the cells were suspended in five pellet volumes of 0.3 M sucrose and 2% Tween 40 in Buffer A [10 mM HEPES-KOH, pH 7.9, containing 10 mM KCl, 1.5 mM MgCl2, 0.1 mM EGTA, 0.5 mM DTT, 0.5 mM PMSF, 2 µg/ml leupeptin, and 2 µg/ml pepstatin A]. After freezing, thawing, and gently homogenizing, nuclei were isolated by centrifugation at 25,000 x g on a 1.5 M sucrose cushion containing the same buffer. Nuclei were lysed in Buffer B [10 mM HEPES-KOH, pH 7.9, 420 mM NaCl, 1.5 mM MgCl2, 0.1 mM EGTA, 10% glycerol, 0.5 mM DTT, 0.5 mM PMSF, 2 µg/ml leupeptin, and 2 µg/ml pepstatin A]. After centrifugation at 100,000 x g for 1 h, the supernatant was dialyzed for use in gel mobility shift analyses using 10 mM Tris-Cl at pH 7.9, 1 mM MgCl2, 1 mM DTT, 1 mM EDTA, and 5% glycerol, aliquoted, and stored at -70 C.
EMSAs
Oligonucleotides used for EMSA were synthesized or were purified
from 2% agarose gel using QIAEX (Quiagen, Chatsworth, CA) after
restriction enzyme treatment of the chimeric CAT constructs. They were
labeled with [-32P]dCTP using Klenow or with
[-32P]ATP using T4 polynucleotide kinase and purified
on an 8% native polyacrylamide gel (28, 47, 61). EMSAs were performed
as previously described (28, 47) but used two different binding
conditions. Binding reactions in low salts and the absence of
detergents were carried out in a volume of 20 µl for 30 min at room
temperature. The reaction mixtures contained 1.5 fmol
[32P]DNA, 2 µg cell or nuclear extract, and 0.5 µg
poly(dI-dC) in 10 mM Tris-Cl at pH 7.9, 1 mM
MgCl2, 1 mM DTT, 1 mM EDTA, and 5%
glycerol. Alternatively, binding reactions were performed in high salts
plus detergent and included 1.5 fmol of [32P]DNA, 2 µg
nuclear extract, and 0.5 µg poly(dI-dC) in 10 mM Tris-Cl
at pH 7.9, 5 mM MgCl2, 50 mM KCl, 1
mM DTT, 1 mM EDTA, 0.1% Triton X-100, and
12.5% glycerol. Where indicated, unlabeled oligonucleotides were added
to the binding reaction as competitors and incubated with the extract
for 20 min before the addition of labeled DNA. In experiments using
antiserum, extracts were incubated in the same buffer containing
antiserum or normal rabbit serum at 20 C for 1 h before being
processed. After incubations, reaction mixes were subjected to
electrophoresis on 5% native polyacrylamide gels at 160 V in
0.5xTris-borate-EDTA and at room temperature. Gels were dried and
autoradiographed.
Other Assays
Protein concentration was determined by Bradford’s method
(Bio-Rad) and used recrystallized BSA as the standard.
Statistical Significance
All experiments were repeated at least three times with
different batches of cells. Values are the mean ± SE
of these experiments where noted. Significance between experimental
values was determined by two-way ANOVA and are significant if
P values were <0.05 when data from all experiments were
considered.
|
FOOTNOTES |
|---|
V. M., C. G., and M. S. were supported by the Interthyr Research Foundation, 301 St. Paul Place, Suite 712, Baltimore, MD 21202, during this project. M. S. was also supported by a Thyroid Research Council Award from the Knoll Pharmaceutical Company, 3000 Continental Drive-North, Mount Olive, NJ 07828, during a portion of this project.
1 Current Address: 1st Department of Internal Medicine, Tottori
University School of Medicine, Yonago 683, Japan. 
2 Current address: Department of Internal Medicine, Chungnam
National University Hospital, 640 Daesa-Dong, Chung-ku, Daejon,
301–040, Korea.
3 Current address: Cattedra di Endocrinologia, Università degli
Studi "G. D’Annunzio" - Chieti, Faculty of Medicine and Surgery,
Palazzina Scuole di Specializzazione, Via dei Vestini, 66100 Chieti,
Italy.
4 Current address: Department of Surgery, Johns Hopkins University,
Ross Building, Room 756, 720 Rutland Avenue, Baltimore, Maryland
21287.
Received for publication April 18, 1997. Revision received October 21, 1997. Accepted for publication October 23, 1997.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
