| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Gene Expression in Thyrocytes by Single Strand Binding Protein-1, a Transcription Factor That Also Regulates Thyrotropin Receptor and MHC Class I Gene Expression
Cell Regulation Section, Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases (P.L.B.-S., K.S., M.Oht., J.S., M.Ohm., V.M., G.N., M.S., S.-I.T., M.P., S.L., A.M., L.D.K.), and Experimental Immunology Branch, National Cancer Institute (D.S.S.), National Institutes of Health, Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Dr. Leonard D. Kohn, Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, Building 10, Room 9C101B, National Institutes of Health, Bethesda, Maryland 20892-1360. E-mail: lenk{at}bdg10.niddk.nih.gov
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(IFN) to
overcome the normally suppressed state of the gene and induce aberrant
class II expression. In this report we show that SSBP-1, when
overexpressed in FRTL-5 thyroid cells, is a positive regulator of human
leukocyte antigen (HLA)-DR
class II gene expression, as is IFN or
the class II trans-activator (CIITA). This is evidenced
by increased exogenous promoter activity, increased endogenous RNA
levels, and increased endogenous antigen expression after transfecting
full-length SSBP-1 complementary DNA together with a HLA-DR
promoter-reporter gene chimera into TSH-treated FRTL-5 thyroid cells
whose endogenous SSBP-1 levels are low. IFN reverses the ability of
TSH to decrease endogenous SSBP-1 RNA levels. Also, whereas SSBP-1
transfection does not cause any increase in IFN-induced exogenous
promoter activity, transfection of SSBP-1 and CIITA additively
increases endogenous class II RNA levels to levels measured in cells
treated with IFN. Further, competition studies show that SSBP-1
binding is necessary for formation of the double strand protein/DNA
complexes that are seen in electrophoretic mobility shift assays when
the class II 5'-flanking region is incubated with extracts from
IFN-treated FRTL-5 cells and that have been previously associated
with IFN-induced aberrant class II expression. These data suggest
that SSBP-1 is involved in the action of IFN to overcome the
normally suppressed state of the class II gene; it functions together
with CIITA, whose expression is independently increased by IFN. The
effect of SSBP-1 as a positive regulator of class II promoter activity
is lost in cells maintained without TSH, in which endogenous SSBP-1 RNA
levels are already high in the absence of aberrant class II gene
expression. These data suggest that high levels of endogenous SSBP-1
are insufficient to cause aberrant class II expression, but, rather,
TSH or IFN treatment additionally modulates the cell, albeit
differently, such that transfected or endogenous SSBP-1, respectively,
can express its positive regulatory activity. The effect of TSH is
consistent with reports indicating that TSH enhances the ability of
IFN to increase class II gene expression despite the fact IFN
increases endogenous SSBP-1 to only the same levels as in cells
untreated with TSH. Finally, the effect of SSBP-1 as a positive
regulator is lost when GXXXXG motifs, which exist on both the coding
and noncoding strands of the S box, are mutated. Consistent with this,
mutation and oligonucleotide competition studies show that GXXXXG
motifs are necessary for either strand of the S box to bind protein/DNA
complexes containing SSBP-1 in FRTL-5 cell extracts or to bind to
recombinant SSBP-1. They also suggest that the SSBP-1-binding sites on
either strand of the HLA-DR S box are functionally distinct. We
conclude from these data that the positive regulatory action of SSBP-1
on class II gene expression involves GXXXXG motifs on each strand of
the highly conserved S box of the class II 5'-flanking region. As
SSBP-1 is modulated by IFN and is involved in class I and TSH
receptor as well as class II gene expression in FRTL-5 cells, the sum
of the data supports the hypotheses that common transcription factors
regulate all three genes, and their altered activities may contribute
to the development of autoimmunity. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MHC class I, like class II molecules, function in immune surveillance and self-recognition by binding and presenting intracellularly derived peptides to the immune system; unlike class II, class I is normally expressed on thyrocytes (5, 7, 14). In the rat FRTL-5 cell, TSH and the hormones necessary for growth and function of the cell decrease MHC class I gene expression coordinately with decreased TSHR gene expression (5, 15, 16, 17, 18). As abnormally elevated class I (10, 12) and aberrant class II expression are associated with ATD (9, 10, 11, 12), and suppression of class I and class II genes by iodide and/or methimazole is associated with their therapeutic effects in ATD and/or other immune disorders (5, 12, 19, 20), we developed the following hypothesis (5, 15, 16, 17, 18, 21, 22). Expression of MHC class I and TSHR genes in the normal thyrocyte is suppressed by TSH or other hormones. This allows the ordered progression of the cell cycle and prevents triggering immune responses to gene products that increase during the growth and functional phases of the cell cycle. To achieve this, common transcription factors coordinately suppress both genes, thereby allowing the cross-talk necessary for self-tolerance during TSHR-mediated changes in function and growth. One or more of these transcription factors also regulates MHC class II expression. Coordinate control of MHC genes by these transcription factors is important to maintain tolerance and prevent autoimmunity, as the MHC genes function together in immune surveillance (7), coordinately increase in ATD (10, 12), and coordinately decrease during treatment of ATD (12).
One of the transcription factors that coordinately regulates TSHR and
class I gene expression is a single stranded DNA-binding protein
(SSBP-1) (18, 23, 24). In the course of our studies identifying SSBP-1
and its role in the TSHR (24), we noted that a GXXXXG motif, which is a
critical component of SSBP-1-binding sites on the TSHR, was present in
the S box of MHC class II genes. The S box is one of four highly
conserved elements, S, X1, X2, and Y, necessary
for constitutive class II expression in lymphocytes and IFN-induced,
aberrant human leukocyte antigen (HLA)-DR gene expression both in
antigen-presenting cells of the immune system and in the FRTL-5
thyrocyte (7, 8, 21, 22, 25, 26). We showed that an oligonucleotide
mimicking the sequence of the mouse class II S box could compete for
SSBP-1 binding to the TSHR SSBP-1 site and that recombinant SSBP-1
could bind to the mouse class II S box oligonucleotide (24). These data
suggested that SSBP-1 might be one of the transcription factors
exerting coordinate control of class II as well as TSHR and class I
gene expression in thyrocytes. The present report further evaluates
this possibility.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
was obtained from Life Technologies
(Grand Island, NY). [-32P]Deoxy-CTP (3000 Ci/mmol) and
[14C]chloramphenicol (50 mCi/mmol) were purchased from
DuPont-New England Nuclear (Boston, MA), [-32P]ATP
(6000 Ci/mmol) was obtained from Amersham (Arlington Height, IL).
Synthetic oligonucleotides were purchased from Operon Technologies
(Alameda, CA). Antiserum directed at activating transcription factor-1
(ATF-1) was obtained from Santa Cruz Biotechnology (Santa Cruz, CA);
antiserum to cAMP response element-binding protein-327 (CREB-327) (27)
and its control counterpart were gifts from Dr. James P. Hoeffler,
Invitrogen (San Diego, CA). All other materials were obtained from
Sigma Chemical Co. (St. Louis, MO) unless otherwise noted.
Cell culture
FRTL-5 rat thyroid cells (Interthyr Research Foundation,
Baltimore, MD; American Type Culture Collection CRL8305) were a fresh
subclone (F1) with the properties previously described (15, 16, 17, 18, 21, 22, 23, 24, 28, 29). They were grown in Coon’s modified F-12 medium containing 5%
heat-treated, mycoplasma-free calf serum, 1 mM nonessential
amino acids, and a mixture of six hormones (6H) containing 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). Cells were diploid and between their 5th
and 25th passage. Fresh medium was added every 2 or 3 days, and cells
were passaged every 7–10 days. In some experiments, cells were grown
to near confluence in 6H medium, then maintained in 5H medium, which
contains no TSH, for 6 days before starting the experiment.
Plasmids
The HLA-DR promoter constructs coupled to the chloramphenicol
acetyltransferase (CAT) reporter gene were provided by Dr. L. H.
Glimcher and A. M. Reimold of the Harvard School of Public Health
and Department of Medicine, Harvard University Medical School (Boston,
MA). Their construction and characteristics were previously described
(21, 22, 30). An additional CAT plasmid containing the mutated S box in
the 5'-flanking region of HLA-DR was constructed by PCR, using
the HLA-DR-CAT chimera containing -137 to +45 bp of 5'-flanking
region as template and the following primers: a 5'-primer with a
5'-SphI restriction site,
5'-CGGAGTGCATGCTGTCCTAGACCTCTTGCAAGAAC-3', and a 3'-primer with an
XbaI restriction site, 5'-TTGGGCTCTAGATTGGGAGTCAGTAGAGCT-3'
(21, 22, 30). The PCR products were purified by phenol/chloroform
extraction and ethanol precipitation, digested with XbaI and
SphI, purified from a 2% agarose gel with Jet Sorb
(Genomed, Frederick, MD), and religated (DNA ligation kit, Takara
Biochemical, Berkeley, CA) into the pCAT-Basic Vector (Promega,
Madison, WI) at the SphI and XbaI sites as
previously described (21, 22, 30).
The SSBP-1 expression vector, pcDNA3-SSBP-1, was constructed by ligating the rat SSBP-1 sequence with the pcDNA3 vector (24). pcDNA3-CIITA was the plasmid used previously (21) and is the CIITA-8 clone containing the full-length complementary DNA (cDNA) plus an additional 20 bp of 5'- and 1.6 kb of 3'-untranslated sequence (31, 32). pRSV-luciferase, which is used to measure the efficiency of transfection, was provided by Dr. S. Subramani, University of California (La Jolla, CA) (33). DNA was prepared and purified by CsCl gradient centrifugation (34). The sequences of all constructs were confirmed by a standard method (35).
Transient expression analysis
Transient transfections in FRTL-5 cells were performed as
previously described (17, 18, 21, 22), using one of the following
procedures. In the first, cells were cultivated in 6H medium to
approximately 80% confluence, harvested, washed, and resuspended
(1.5 x 107 cells/ml) in 0.85 ml electroporation
buffer [272 mM sucrose, 7 mM sodium phosphate
buffer (pH 7.4), and 1 mM MgCl2]. Plasmid DNA
was added (20 µg of the CAT chimera together with 2 µg
pRSV-luciferase); cotransfections with pcDNA3-SSBP-1 or its vector
control, pcDNA3, also used 20 µg DNA. Cells were pulsed (300 V;
capacitance, 960 µF), plated (6 x 106 cells/10-cm
dish), and cultured in 6H medium plus 5% calf serum with or without
100 U/ml IFN. At the times noted, cells were harvested for CAT and
luciferase assays. The second procedure differed as follows. After
FRTL-5 cells were grown to 80% confluence in 6H medium, they were
maintained for 6 days in 5H medium plus 5% calf serum. Cells were
returned to 6H medium for 12 h and transfected as described above.
Twelve hours later, the medium was changed to fresh 5H medium with 5%
calf serum with or without IFN. Cell viability was approximately
80% in all experiments.
CAT activity was measured as previously described (17, 18, 21, 22, 36), using 10–30 µg cell lysate in a final volume of 130 µl. Incubation was performed at 37 C for 4 h with acetyl coenzyme A supplementation (20 µl of a 3.5 mg/ml solution) after 2 h. Acetylated chloramphenicol was separated by TLC and autoradiographed; positive spots were cut and quantitated in a scintillation spectrometer. CAT values were normalized to luciferase activity using the Promega assay system and a Berthold luminometer (Wallac, Inc., Gaithersburg, MD).
Cellular extracts
Cell extracts were made by a modification (17, 18, 21, 22) of
the method of Dignam et al. (37). Briefly, FRTL-5 cells were
harvested by scraping after being washed twice in ice-cold PBS and
pelleted. The pellet was resuspended in 2 vol Dignam buffer C [20
mM HEPES (pH 7.9), 1.5 mM MgCl2,
0.42 M NaCl, 25% glycerol, 0.5 mM
dithiothreitol (DTT), 0.5 mM phenylmethylsulfonylfluoride,
1 µg/ml leupeptin, and 1 µg/ml pepstatin]. The final NaCl
concentration was adjusted on the basis of cell pellet volume to 0.42
M. Cells were lysed by repeated cycles of freezing and
thawing. The extracts were centrifuged at 100,000 x g
at 4 C for 20 min. The supernatant was recovered, aliquoted, and stored
at -70 C.
Electrophoretic mobility shift assays (EMSAs)
Oligonucleotides used for EMSA were synthesized or purified from
2% agarose gel using QIAEX (Qiagen, Chatsworth, CA) or Jet-Sorb
(Genomed), after restriction enzyme treatment of the chimeric CAT
constructs described above (21, 22. 30). They were labeled with
[-32P]deoxy-CTP using Klenow or with
[-32P]ATP using T4 polynucleotide kinase, then
purified on an 8% native polyacrylamide gel (17, 18, 21, 22, 38).
EMSA were performed basically as previously described (17, 18, 21, 22). Binding reactions were carried out in a volume of 20 µl for 30 min at room temperature. The reaction mixtures contained 1.5 fmol [32P]DNA, 3 µg cell extract, and 1.5 µg poly(dI-dC) in 10 mM Tris-Cl (pH 7.9), 1 mM MgCl2, 1 mM DTT, 1 mM EDTA, and 5% glycerol. Where indicated, unlabeled double or single stranded oligonucleotides were added to the binding reaction as competitors and incubated with the extract for 20 min at room temperature before the addition of labeled DNA. Similarly, in experiments using antiserum, extracts were incubated in the same buffer containing antiserum or normal rabbit serum for 20 min at room temperature before being processed as above. After incubations, reaction mixtures were subjected to electrophoresis on 3.5% or 5% native polyacrylamide gels at 160 V in 1.0 x Tris-borate-EDTA at room temperature for 1.5–2 h. Gels were dried and autoradiographed at -80 C overnight unless otherwise noted.
Protein production in Escherichia coli
Recombinant SSBP-1 protein was produced using the pET system
(Novagen, Madison, WI). SSBP-1 cDNA was ligated to the EcoRI
site of the expression vector, pET-30(+), allowing the His-Tag sequence
to be linked to its N-terminus. After transformation using E.
coli BL21 (DE3), a single colony was inoculated in 50 ml Luria
Bertoni medium containing 30 µg/ml kanamycin and incubated with
shaking at 37 C. At 0.6 OD600,
isopropyl-ß-D-thiogalactopyranoside was added to 1
mM. After 2 h, the induced cells were collected by
centrifugation (5,000 x g, 5 min, 4 C), resuspended in
4 ml ice-cold binding buffer (5 mM imidazole, 0.5
M NaCl, and 20 mM Tris-HCl, pH 7.9), then
sonicated until no longer viscous. Cell extracts were centrifuged
(39,000 x g, 20 min, 4 C), the supernatant was applied
to His-Bind columns containing resin-immobilized Ni2+, and
the columns were washed with 25 ml binding buffer. Unbound proteins
were removed with 15 ml wash buffer (60 mM imidazole, 0.5
M NaCl, and 20 mM Tris-HCl, pH 7.9); SSBP-1 was
recovered with 15 ml elute buffer containing 1 M imidazole.
The His-Bind column contained 5 ml resin and was washed, sequentially,
with 7.5 ml deionized water, 12.5 ml charge buffer (50 mM
NiSO4), and 12.5 ml binding buffer. After the addition of a
one third volume of strip buffer (100 mM EDTA, 0.5
M NaCl, and 20 mM Tris-HCl, pH 7.9), the eluted
fraction was dialyzed against 20 mM HEPES-KOH (pH 7.9), 100
mM KCl, 0.1 mM EDTA, 20% glycerol, 0.5
mM DTT, 0.5 mM phenylmethylsulfonylfluoride, 2
µg/ml leupeptin, and 2 µg/ml pepstatin A, then concentrated in a
Centricon 10 (Amicon, Beverly, MA) for use in the EMSA.
Other methods
The protein concentration was determined by Bradford’s method
(Bio-Rad, Richmond, CA); recrystallized BSA was the standard.
Total cellular RNA was isolated, and Northern analyses were performed
as described (39). The SSBP-1 probe was the clone 9 insert used
previously (24); rat ß-actin was provided by Dr. B. Paterson (NCI,
Bethesda, MD). Radiolabeling of these probes was previously described
(39). The MHC class II DNA probe was a PCR-amplified 546-bp product
from between 74–619 bp of the class II sequence. It was obtained using
RNA from IFN-treated rat FRTL-5 cells, a sense primer with the
nucleotide sequence 5'-AGCAAGCCAGTCACAGAAGG-3', and an antisense primer
with the sequence 5'-GATTCGACTTGGAAGATGCC-3'. Both primer regions are
highly conserved in the class II nucleotide and protein sequences.
After amplification using Pfu DNA polymerase, the product was purified
on agarose gels and then random prime labeled using
[32P]deoxy-CTP and Klenow enzyme. Hybridization (1.0
x 106 cpm/ml) and washing were performed as previously
described (39).
For fluorescence-activated cell sorter (FACS) analysis, single cell suspensions were prepared and stained as previously described (17, 40), except that HLA-DR antigen was detected using a class II-specific murine monoclonal antibody (22, 41); Leu-4 was used as a background control.
Statistical significance
All experiments were repeated at least three times with
different batches of cells. Values are the mean ±SD of
these experiments where noted. Significance between experimental values
was determined by two-way ANOVA and are significant at P
[lt ]0.05 when data from all experiments were considered.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene expression in FRTL-5
thyroid cells: its effect requires an intact S box containing a GXXXXG
motif on the coding strand
) was found to be critical for
both binding and function. In that study, we noted that the S box of
the mouse MHC class II gene (Fig. 1A, underlined) had
multiple GXXXXG motifs on both the coding and noncoding strands. We
observed that the radiolabeled noncoding strand encompassing the mouse
S box, -119 to -88 bp, could bind recombinant SSBP-1, as did the
radiolabeled noncoding strand oligonucleotide of the TSHR. We also
observed that unlabeled TSHR or mouse class II S box noncoding strand
oligonucleotides were able to prevent the binding of either
radiolabeled probe to recombinant SSBP-1 or to the FRTL-5 cell nuclear
extracts (24). This raised the possibility that SSBP-1 might be
functionally involved in regulating MHC class II expression in FRTL-5
cells.
|
-137 bp CAT chimera, as
described in previous studies (21, 22). FRTL-5 thyroid cells maintained
in medium with no TSH have high SSBP-1 RNA levels and maximal SSBP-1
complex formation with the TSHR; TSH treatment reduces both (24). We,
therefore, cotransfected the full-length SSBP-1 cDNA with the -137 bp
HLA-DR-CAT chimera into cells maintained in the presence of TSH to
decrease endogenous levels of the SSBP-1 transcription factor (Fig. 1C
).
The HLA-DR-CAT chimera is not normally expressed in transiently
transfected FRTL-5 thyrocytes maintained in the presence of TSH,
compared to the vector control, unless the cells are treated with
recombinant rat IFN (Fig. 1C) (21, 22). Cotransfection of SSBP-1,
however, caused a significant increase in basal promoter activity,
albeit less than IFN, whereas cotransfection of pcDNA3, the vector
control, had no positive regulatory effect (Fig. 1C) despite equal
transfection efficiencies.
The HLA-DR 5'-flanking region has two GXXXXG sequences on its
noncoding strand, each of which overlaps the minimal S box sequence
(Fig. 1B, solid line). The coding strand has another GXXXXG
sequence that is immediately 5' to the minimal S box motif, but is
encompassed within a broader S box motif defined by some workers in the
field (Fig. 1B, dashed underline) (7, 25, 26, 30). We
mutated nucleotides to eliminate the GXXXXG motif on the coding strand
and to eliminate one GXXXXG motif on the noncoding strand (Fig. 1B).
The reason for mutating GXXXXG sites on both strands will become
evident below in the studies that show that the GXXXXG motif on each
strand can bind recombinant SSBP-1. Just as mutation of the GXXXXG site
in the TSHR results in loss of SSBP-1 positive regulation of the TSHR
(24), mutation of the GXXXXG sites in the HLA-DR S box resulted in a
loss of SSBP-1 up-regulation of the class II gene in FRTL-5 cells. Also
notable was the observation that the mutation decreased the ability of
IFN to induce aberrant class II expression (Fig. 1C).
The ability of transfected SSBP-1 to increase exogenous HLA-DR gene
expression was paralleled by an increase in endogenous class II RNA
levels and by increased class II antigen expression on the cell surface
(Fig. 2). Thus, transfection of SSBP-1,
like transfection of the class II trans-activator (CIITA),
an intermediate in the action of IFN in FRTL-5 cells (21), or
treatment of cells with IFN itself, increased class II RNA levels as
measured in Northern analyses (Fig. 2A, lanes 2–4, respectively,
vs. lane 1). The increase was evident in blots in which each
sample contained equal amounts of RNA (Fig. 2A, lanes 8–11); the
increase was not associated with a change in ß-actin RNA levels or
RNA yield (data not shown). The SSBP-1-induced increase in antigen
expression on the surface of the cells was measured by FACS analysis
(Fig. 2B).
|
, to increase
both endogenous class II RNA levels and antigen expression (Fig. 2, A
and B, respectively) is consistent with its lower activity in
experiments measuring exogenous HLA-DR promoter activity (Fig. 1C).
The lesser effect in both experiments (Figs. 1C and 2) may simply
reflect the fact that transient transfection by the procedure used
herein would not be expected to involve all cells. Additionally or
alternatively, it may reflect the fact that SSBP-1 positive regulatory
activity is only one component of a complex process leading to aberrant
class II expression. This is evident in the following experiments.
Cotransfection of SSBP-1 plus CIITA, each at maximally effective
plasmid concentrations, as determined in preliminary experiments (data
not shown), increased endogenous class II RNA levels more than
transfection of either alone (Fig. 2A, lane 6 vs. lanes 5 or
7). Moreover, the increase effected by cotransfection of both was
similar to levels induced by IFN (Fig. 2A, lane 4). This was a
specific effect that did not reflect changes in RNA loading (Fig. 2A, lanes 12–14) or altered ß-actin RNA levels (data not shown). As
there was no significant difference (±7%) in transfection efficiency
using SSBP-1, CIITA, or both in these experiments, these data raised
the possibility that the mechanisms by which SSBP-1 and CIITA increased
class II were not identical, but, rather, additive. Each change was a
component of the complex series of events involved in the
IFN-induced increases in class II expression. This possibility was
supported by the next experiments. Whereas TSH treatment of FRTL-5
cells decreased SSBP-1 RNA levels (Fig. 3) (24) and complex formation with the
TSHR and mouse S box binding sites (24), IFN reversed the ability of
TSH to decrease SSBP-1 RNA levels (Fig. 3) and complex formation (data
not shown).
|
(Fig. 4A), in which endogenous SSBP-1 levels
are high and class II promoter activity is maximally increased (Fig. 3), even in TSH-treated cells. In sum, IFN not only increases the
levels of a transcription factor, SSBP-1, which is a positive regulator
of class II expression, but also induces the expression of the CIITA
coregulator that is necessary for aberrant expression, but does not
bind to the DNA (21). Therefore, whereas increases in SSBP-1 and CIITA
appear to be additive in their ability to increase class II transcripts
(Fig. 2A), SSBP-1 will not additively perturb IFN-increased class II
expression.
|
class II
gene expression in FRTL-5 thyrocytes is lost in cells maintained in the
absence of TSH (Fig. 4, B vs. A, stippled bars).
Additionally, IFN does not alter SSBP-1 RNA levels (Fig. 3, open bars) or complex formation (data not shown) in cells
maintained without TSH, in which levels already are presumed to be
maximal. We suggest that this is not simply because endogenous SSBP-1
levels are high in cells maintained without TSH compared to those in
cells treated with TSH (Fig. 3). Rather, we suggest that TSH may
contribute other effects that are necessary to express the SSBP-1
positive regulatory action. This would be consistent with the ability
of TSH to enhance IFN-increased exogenous HLA-DR promoter
activity and endogenous antigen expression in rat thyroid cells (21, 22, 42). As will be discussed (see Fig. 9 and associated text), we
suggest that this would be consistent with the ability of TSH/kinase A
to coordinately activate a suppressor factor, TSHR suppressor
element-binding protein-1, termed TSEP-1. Activated TSEP-1, a homolog
of the human Y box protein YB-1, can induce single strand or triple
helix formation in the S box region, facilitate SSBP-1 single strand
binding, and thereby allow the positive regulatory action of SSBP-1 to
be expressed when TSH-treated cells are transfected with SSBP-1.
Certainly, the TSH action could well involve other effects.
Nevertheless, it seems clear that high levels of SSBP-1 alone are
insufficient for increased class II gene expression in a normally
regulated cell or in the absence of CIITA.
|
-increased HLA-DR gene expressions
in FRTL-5 thyroid cells gene forms a major
multicomponent protein/DNA complex when incubated with extracts from
cells maintained in the absence of IFN and when evaluated in EMSAs
(Fig. 5A, lane 2, basal or A complex)
(21, 22). IFN treatment of FRTL-5 cells induces the formation of a
faster migrating complex (Fig. 5A, lane 1, complex B) (21, 22) and
enhances the formation of a complex with the migratory properties of
the basal complex (Fig. 5A, lane 1, complex A) (21, 22). In both the
presence of TSH (Fig. 4B, lanes 3 and 4) and its absence (Fig. 5B, lanes 1 and 2), a single strand oligonucleotide with the sequence of
the upstream SSBP-1-binding site [USSSBPE(-)] on the noncoding
strand of the TSHR (Fig. 5, bottom) decreases the formation
of both the upper and lower complexes (Fig. 5B, lane 2 vs. 1
and lane 4 vs. 3). The inhibitory effect is more evident in
TSH-treated cells (Fig. 5B, lane 4 vs. 2), however, where
endogenous SSBP-1 levels are lower (Fig. 3). Cells were transfected
with pcDNA3 in these and subsequent experiments to ensure that the gel
shifts were performed under conditions comparable to the transfections
in Figs. 1 and 4. These data suggested that SSBP-1 is a component of
both the A and B complexes; the following data support this and further
define the SSBP-1-binding sites on the class II S box.
|
, lane 5
vs. lanes 1–4) by the counterpart oligonucleotide with the
coding strand sequence of the upstream site [USSSBPE(+)]. USSSBPE(+)
has no SSBP-1 site or GXXXXG motif, as evident from the noncoding
strand sequence in Fig. 1A. Additionally, the ability of the upstream
SSBP-1-binding element on the TSHR, USSSBPE(-), to inhibit the
formation of both complexes was lost if the GXXXXG motif was mutated
(Fig. 6A, lane 6).
|
), and both appeared
to be the same complexes characterized in our separate reports
describing the action of IFN or CIITA on these complexes (21, 22).
Thus, as previously evidenced (21, 22), a dominant component of the A
complex is ATF-1, as evidenced by the ability of a specific antiserum
to ATF-1 to inhibit its formation (Fig. 6C, lane 1 vs. 2).
Also, as previously demonstrated (21, 22), formation of the lower,
faster migrating B complex induced by IFN was inhibited by a double
strand oligonucleotide containing the Y box, -87 to -55 bp (Fig. 6D),
but not its single strand components (data not shown). These data
suggested that SSBP-1, which interacted with the noncoding strand of
the TSHR, was a component of both complexes associated with
IFN-increased MHC class II expression in FRTL-5 thyroid cells (21, 22): the complex enhanced by IFN treatment (complex A) and the
faster migrating complex induced by IFN (complex B).
We could not, in our previous reports (21, 22), demonstrate the
presence of CREB in the class II complexes, using specific antisera to
CREB-1 or CREB-2 from Santa Cruz despite the fact the X2
box is homologous to a CRE (7, 8, 25, 26) and that an CREB/ATF-binding
site is suggested to be required for aberrant expression of the MHC
class II DR promoter in T antigen-transformed COS cells (43).
However, using a different CREB antiserum in these studies,
anti-CREB-327 (27), we showed that the multicomponent complexes formed
between the 5'-flanking region of the HLA-DR gene and cell extracts
from IFN-treated FRTL-5 thyroid cells might contain CREB. This is
evidenced by an alteration in their appearance when anti-CREB-327 was
present and by the formation of a new complex, not evident when a
control serum was added to the incubation (Fig. 6C, lane 4
vs. 3). The anti-CREB-327 data further emphasize the
specificity of the anti-ATF-1 data and the multicomponent nature of the
complexes.
Whereas SSBP-1 binding was specific for the noncoding strand of the
TSHR, this was not the case for the HLA-DR class II S box. Thus,
using extracts from IFN-treated FRTL-5 cells, we could see complexes
formed with coding (+) as well as the noncoding (-) strands of the
mouse or HLA-DR S box (Fig. 7, A and
B, respectively). Whereas the mouse S box noncoding (-) strand
appeared to form complexes with the extracts better than the mouse S
box coding (+) strand, the opposite was true for HLA-DR at
comparable amounts of labeled probe (Fig. 7A, lane 2 vs. 6).
The HLA-DR S box coding (+) strand appeared to form complexes better
than the HLA-DR S box noncoding (-) strand (Fig. 7B, lane 2
vs. 6). Additionally, the HLA-DR coding (+) strand was a
better unlabeled competitor (Fig. 7, A and B, lane 3 vs. 4
and lane 7 vs. 8 in both). Nevertheless, the GXXXXG motif
was important for the ability of each strand of the HLA-DR S box to
bind SSBP-1. Mutation of the GXXXXG sites on each strand, as detailed
in Fig. 1B, not only eliminated the functional activity of SSBP-1 (Fig. 1C), but also eliminated the ability of each strand to prevent complex
formation with extracts from cells treated with IFN (Fig. 7C, lane 3
vs. 4 and lane 5 vs. 6).
|
, lane 3).
The same complexes in this experiment were inhibited by the S box on
the HLA-DR coding (+) strand (Fig. 8A, lane 1). More importantly,
recombinant SSBP-1 was able to bind to either the HLA-DR S box
coding (+) or noncoding (-) strand (Fig. 8B, lanes 4 and 8,
respectively), and in each case, the recombinant SSBP-1 complex(es)
migrated with a mobility similar to that of a complex formed with the
FRTL-5 cell extracts (Fig. 8B, lanes 4 vs. 1 and lane 8
vs. 5, respectively). The binding of recombinant SSBP-1 to
each strand was specific, as evidenced by self competition (Fig. 8B, lanes 2 and 6, respectively). Additionally, like the extracts, the
complexes formed with each radiolabeled strand S box oligonucleotide
were different. Thus, consistent with the existence of differences in
the S box binding site on each strand, SSBP-1 binding to the
radiolabeled HLA-DR coding (+) strand S box was inhibited better by
its unlabeled counterpart than that to the unlabeled noncoding (-)
strand S box (Fig. 8B, lane 2 vs. 3). Conversely, SSBP-1
binding to the radiolabeled HLA-DR noncoding (-) strand S box was
inhibited better by the unlabeled noncoding (-) strand than by the
unlabeled coding (+) strand S box oligonucleotide (Fig. 8B, lane 6
vs. 7).
|
), when recombinant SSBP-1 and
FRTL-5 cell extracts were mixed, an additional complex, termed X, was
formed. Also, the appearance of the complexes formed with only the cell
extracts was altered (Fig. 8C, lane 3 vs. lanes 2 and
4).
In sum, these data suggest that each strand of the HLA-DR S box can
bind SSBP-1. They indicate that a GXXXXG motif is critical for complex
formation with each strand of the HLA-DR S box. They suggest,
however, that the GXXXXG binding motif is not sufficient for full
binding activity to each strand. Thus, the binding sites on each strand
are distinct, as evidenced in the self-competition experiments (Fig. 7). The mutation data support the view that binding of SSBP-1 to each
strand is functionally relevant, as the same mutations eliminate SSBP-1
positive regulation (Fig. 1).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Certainly these experiments will generate additional questions and more controversy. For example, concerns will be expressed that fibroblasts are not thyroid cells, and the latter, unlike the former, will not act as antigen-presenting cells. Thus, thyroid cells do not express certain well described costimulatory molecules, B7-1 or B7-2 (47, 48); this would prevent their acting as antigen-presenting cells and, instead, result in anergy. However, Kundig, et al. (49) showed that immunization of mice with fibroblasts transfected with a viral protein could induce a cytotoxic T lymphocyte response in the absence of costimulatory molecules on the immunizing fibroblasts. This suggests that aberrant class II on fibroblasts is not different from that on thyrocytes, and increased class II can induce autoimmunity, not anergy. The researchers proposed in that report that costimulatory signals could be host derived (49). Matsuoka et al. also suggest that costimulatory molecules in ATD can be provided by bystander professional antigen-presenting cells (48).
It behooves us, therefore, to focus on increasing our understanding of the complex process involved in the autoimmune response: how aberrant class II might develop, what relationship exists between aberrant class II and abnormal class I in ATD, and what is the complex processing path between aberrant class II and an autoantibody or cytotoxic T lymphocyte response. In this and our recent reports (21, 22), we have focused on identifying factors and elements involved in regulating class II expression in normal thyrocytes, their link to class I regulation, and their relevance to aberrant class II expression.
We have used rat FRTL-5 thyroid cells as our model thyroid system. Rat
FRTL-5 cells are not tumor cells and exhibit most of the properties of
normal rat thyroid cells in vivo (50, 51). They do not
express MHC class II antigens normally (21, 22, 52, 53, 54); however,
IFN can aberrant class II expression in the cells and the changes in
class II induced by IFN mimic those in ATD (10, 21, 22, 52, 53, 54). We
used the human HLA-DR promoter because it is a representative class
II promoter whose expression has been extensively studied,
transcriptional regulation underlies its expression in
antigen-presenting immune and FRTL-5 cells (7, 8, 21, 22, 25, 26, 30),
and previous studies in FRTL-5 cells indicate that the endogenous class
II gene behaves in accordance with the properties of the transfected
HLA-DR promoter constructs used herein (21, 22).
Thus, HLA-DR promoter activity is normally not expressed in FRTL-5
thyroid cells, mimicking the absence of endogenous class II gene
expression. IFN induces HLA-DR promoter expression in FRTL-5
cells, as it does endogenous class II RNA and protein expression (21, 22). Additionally, the same highly conserved elements on all class II
promoters, S, X1, X2, and Y, are required for
IFN to increase MHC class II gene expression in FRTL-5 cells and
antigen-presenting cells of the immune system (7, 8, 21, 22, 25, 26, 55). Extracts from FRTL-5 cells that are not treated with IFN form a
major protein/DNA complex with the 5'-flanking region of the class II
promoter (21, 22) (complex A in Fig. 5). IFN treatment of the cells
enhances the formation of this complex and induces the formation of
another, faster migrating, protein/DNA complex (21, 22) (complex B in
Fig. 5). Like complexes in antigen-presenting cells of the immune
system, they both involve a multiplicity of proteins, not all of which
are known or functionally characterized (6, 7, 8, 21, 22, 25, 26, 55). We
have shown, nevertheless, that IFN induces the formation of CIITA
RNA in FRTL-5 cells (21), as it does in antigen-presenting cells of the
immune system (8, 55), and that CIITA is an intermediate in the action
of IFN to increase class II gene expression in FRTL-5 cells (21).
For example, CIITA duplicates the ability of IFN to enhance the
formation of complex A and induce the formation of complex B, the
novel, faster migrating, protein/DNA complex (21, 22). Thus, it
appears, as a first approximation, that aberrant HLA-DR class II
expression induced by IFN in FRTL-5 cells has key features of
aberrant class II expression induced by IFN in antigen-presenting
immune cells and is reflected by aberrant exogenous class II gene
expression. Moreover, the IFN-induced or enhanced protein/DNA
complexes formed with the class II promoter in FRTL-5 cells are
relevant to aberrant gene expression induced by IFN.
In separate experiments (21), we identified one factor regulating class
I and class II expression in FRTL-5 cells and relevant to
IFN-induced aberrant class II expression: TSEP-1 (56). TSEP-1 is the
rat homolog of the human Y box protein, YB-1 (56). TSEP-1 was cloned in
FRTL-5 cells during studies searching for the mechanism of
TSH/cAMP-induced negative regulation of the TSHR gene (56). It is a
single strand binding protein that we now know is a suppressor of MHC
class I as well as TSHR gene expression and is important for
TSH/cAMP-induced negative regulation of both genes (3, 5, 18, 56). We
showed that TSEP-1 reversed the ability of IFN or CIITA to increase
class II expression and increase the formation of the B complex,
i.e. it was important in returning class II to its normal
state of suppression (21, 22). These data were consistent with the
observation that YB-1 can suppress IFN-induced class II gene
expression in human glioblastoma cells (57). We additionally showed
that IFN decreases TSEP-1 RNA levels, whereas methimazole reverses
this action of IFN (58). We suggested, therefore, that IFN
increases class II expression in FRTL-5 cells not only by its ability
to increase CIITA levels, but also by its suppressive action on TSEP-1,
the net result being aberrant class II expression. Methimazole reversed
the latter effect of IFN, helping to explain its therapeutic role in
ATD.
In the present report we identify another transcription factor that
regulates MHC class II as well as MHC class I and TSHR gene expression.
Thus, we show that SSBP-1 is a positive regulator of class II as well
as TSHR and MHC class I expression (18, 24). We demonstrate this in
TSH-treated cells, in which endogenous SSBP-1 RNA and protein levels
are decreased. We show that SSBP-1 interacts with the S box of the
class II promoter and that SSBP-1 is a component of the protein/DNA
complexes that we have found to be associated with IFN-increased
class II levels in FRTL-5 cells. We show that IFN modulates SSBP-1
RNA levels, reversing the ability of TSH to decrease SSBP-1 RNA levels.
It is reasonable, therefore, to conclude that IFN-treated thyrocytes
simultaneously increase the level of SSBP-1, which is an enhancer
factor; decrease the level of TSEP-1, which is a class II suppressor
factor; and induce the formation of CIITA, which is a
trans-activating factor that binds transcription factors in
a manner that enhances class II gene expression. The net result is
increased or aberrant class II expression.
The ability of SSBP-1 to bind both strands of the S box and the
functional relevance of binding to each remain unclear. Nevertheless,
it is interesting that one critical binding site for TSEP-1/YB-1 on the
class II promoter is a CT-rich region between the S and X boxes (59)
(Fig. 9). TSEP-1/YB-1 is suggested to
induce or stabilize single strand regions 5' to the CRE-like site in
the X boxes, thereby preventing the loading and/or function of other
double strand binding, positive regulatory transcription factors (59)
(Fig. 9). As SSBP-1 is a positive regulatory factor that interacts with
single strand DNA, the following potential and interesting scenario
emerges (Fig. 9). SSBP-1 and TSEP-1/YB-1 would exhibit kinetically
competitive or balanced binding to this region of the class II
promoter, as it may exist in a single strand or triple helix
configuration under normal circumstances. This would result from
TSEP-1/YB-1 activation by TSH/kinase A and binding to the CT-rich
region (56, 59) (Fig. 9). Although TSH activates TSEP-1/YB-1, it
decreases SSBP-1 RNA levels and activity (24, 56). This causes
constitutive expression to be suppressed by minimizing the positive
regulatory activity of SSBP-1 and by preventing the loading of double
strand binding-activating factors (7, 8, 18, 25, 26, 30, 32, 43, 55, 56, 57, 59) (Fig. 9). Residual SSBP-1 binding is proposed, however, to
poise the gene in a functionally responsive state in
antigen-presenting cells, ready to respond to IFN or other
factors important in an immune response. These would increase SSBP-1
levels, as in the case of the transfection experiments presented
herein. In the case of IFN, TSEP-1/YB-1 RNA levels and function
would also decrease, setting the stage for activation by the CIITA
coregulator and full blown aberrant class II expression (Fig. 9). This
could account for the additional role of TSH in SSBP-1 transfections
described herein, but would not necessarily be the only role.
Another reason to raise this kinetic or dynamic model as a possibility
is as follows. Genes that are not functionally expressed in the thyroid
can undergo methylation. As an example, this is the case for TSHR in
FRT cells, which lose thyroid function in association with the loss of
the cell-restricted thyroid transcription factor (TTF-1) (60, 61).
As a result of the methylation, there is no endogenous TSHR gene
expression despite the fact the cells can express an exogenous TSHR
promoter very well. SSBP-1 may, therefore, counterbalance TSEP-1/YB-1
suppression of the class II promoter, allow the cell to maintain a low
level of class II expression, prevent methylation, and therefore allow
full gene expression under cytokine or other stimulation,
i.e. when IFN decreases TSEP-1/YB-1, increases SSBP-1,
and induces CIITA. SSBP-1 may be a necessary factor to maintain a
low level of class II gene expression despite unmeasurable RNA or
antigen expression in many nonimmune cells, as SSBP-1 and TSEP-1/YB-1
are ubiquitous transcription factors (23, 24, 56). The
SSBP-1-TSEP-1/YB-1 relationship will need to be further resolved with
additional mutation studies, and its function in constitutive
expression resolved in other cells as well as in knockout studies.
Understanding how SSBP-1 levels are regulated in functioning thyrocytes
is, nevertheless, more complex than we previously thought (23, 24).
SSBP-1 RNA levels and complex formation with the TSHR are much higher
in functioning FRTL-5 thyroid cells than in nonfunctioning FRT thyroid
cells or buffalo rat liver (BRL) cells (23, 24). The explanation was
that TTF-1, a thyroid-restricted transcription factor, appeared to
enhance SSBP-1 gene expression (24, 60, 61). As TSH decreased TTF-1
levels, this, in turn, decreased SSBP-1 RNA and protein levels and
decreased TSHR as well as class I gene expression (18, 24, 60, 61).
This is a logical means to prevent TSHR from being expressed as an
autoantigen. In the case of the class II gene, the ability of TSH to
decrease SSBP-1 levels may be a mechanism to further attenuate class II
gene expression despite its already low levels, thereby insuring the
absence of any constitutive expression during TSH-induced growth
and function of the cells. Interestingly, however, IFN decreases the
ability of TTF-1 to bind to its recognition site on the TSHR promoter
(62); yet IFN increases, rather than decreases, SSBP-1 RNA levels in
the presence of TSH. This would suggest that IFN, TSH, and TTF-1
have independent abilities to regulate SSBP-1 transcript levels and
that IFN may bypass TSH-induced suppression of SSBP-1 by a mechanism
that does not involve TTF-1. This will require studies at a promoter
level and may be revealing in understanding the means by which TSHR,
class I, and class II can be regulated by a common factor, SSBP-1, yet
have very different constitutive expressions.
We have shown in the TSHR and class I genes that CREB/ATF-1 interacts
with the CRE in the basal state without TSH and that TSH decreases the
CREB/ATF-1 interaction (18, 63, 64, 65). In short, CREB is a positive
regulator of both TSHR and class I gene expression, and negative
regulation is achieved in those genes in part by the decrease in the
CREB/ATF-1 interaction with the CRE of both genes (18, 63, 64, 65). The
anti-ATF-1 and anti-CREB-327 data show that CREB/ATF-1 interacts with
the HLA-DR 5'-flanking region containing the S, X, and Y boxes and
may be part of the multicomponent protein/DNA complexes associated with
basal as well as IFN-induced class II gene expression (Refs. 21 and
22, and this report). This is interesting in two respects. It raises
the possibility that CREB/ATF-1 will regulate class II as well as class
I and TSHR gene expression and is consistent with data showing that a
CREB/ATF binding motif, i.e. the X2 box, is
required for aberrant expression of the MHC class II DR promoter by
simian virus 40 T antigen (43). The role of CREB/ATF-1 in class II
expression in FRTL-5 cells will have to be examined using HLA-DR
constructs with a mutated X2 box in cotransfection
experiments involving both transcription factors.
In sum, the present report identifies another transcription factor,
SSBP-1, that coordinately regulates TSHR, class I, and class II gene
expression in thyrocytes in addition to TSEP-1/YB-1. The hypothesis
remains viable that common transcription factors coordinately regulate
the expression of all three genes and thereby prevent the loss of self
tolerance and the development of autoimmunity in a TSH-stimulated gland
with large increases in potential thyroid autoantigens. It is
reasonable to speculate that aberrant class II expression in ATD and/or
IFN-stimulated glands may involve increased CIITA, decreased
TSEP-1/YB-1, and increased SSBP-1 gene expression.
|
Footnotes |
|---|
2 Current address: Third Department of Internal Medicine,
Yamanashi Medical University, 1110 Tamaho-cho, Nakakoma-gun,
Yamanashi-ken 409–38, Japan.
3 Current address: Cattedra di Endocrinologia, Università
degli Studi "G. D’Annunzio," Faculty of Medicine and Surgery,
Palazzina Scuole di Specializzazione, Via dei Vestini, 66100 Chieti,
Italy.
4 Supported in part by the Interthyr Research Foundation (Baltimore,
MD) during this project.
5 Current address: Department of Internal Medicine, Chungnam
National University Hospital, 640 Daesa-Dong, Chung-ku, Daejon 301-040,
Korea.
6 Current address: First Department of Internal Medicine, Tottori
University School of Medicine, Yonago 683, Japan.
Received September 18, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B. J Biol Chem 270:11453–11462 gene is
mediated by conserved proximal promoter elements. J Immunol 151:4173–4182[Abstract]
/ß interferon treatment. Immunogenetics 30:18–26[CrossRef][Medline]
-interferon. Endocrinology 121:2087–2092
promoter and activation by SV40 T-antigen. Nucleic Acids Res 20:4881–4887 and non- interferons on the expression of Ia antigen,
growth, and differentiated functions of FRTL-5 cells. Endocrinology 123:2849–2855 interferon and tumor necrosis factor on thyroid
cells: induction of class II antigen and inhibition of growth
stimulation. Mol Cell Endocrinol 58:329–336
induces HLA-DR expression by
thyroid epithelium. Clin Exp Immunol 61:265–273[Medline]
activation of class II major
histocompatibility genes. J Exp Med 179:1605–1611 suppresses thyrotropin receptor promoter activity
by reducing thyroid transcription factor-1 (TTF-1) binding to its
recognition site. Mol Endocrinol 10:826–836This article has been cited by other articles:
 |
|
 |
|
  Y. Imamura, Y. Fujigaki, Y. Oomori, K. Ouryouji, S. Yanagisawa, H. Miyazawa, and P.-L. Wang Transcriptional Regulation of the Salivary Histatin Gene: Finding of a Strong Positive Regulatory Element and Its Binding Protein J. Biochem., March 1, 2009; 145(3): 279 - 288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Harton and J. P.-Y. Ting Class II Transactivator: Mastering the Art of Major Histocompatibility Complex Expression Mol. Cell. Biol., September 1, 2000; 20(17): 6185 - 6194. [Full Text]
|
|
|
|
 |
|
 Z. Wu, P. A. Biro, R. Mirakian, F. Curcio, F. S. Ambesi-Impiombato, and G. F. Bottazzo Transcriptional regulation of the MHC II gene DRA in untransformed human thyrocytes Int. Immunol., April 1, 2000; 12(4): 405 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. Park, H. Kim, J. M. Suh, S. J. Park, O-Y. Kwon, Y. K. Kim, H. K. Ro, B. Y. Cho, J. Chung, and M. Shong Thyrotropin Induces SOCS-1 (Suppressor of Cytokine Signaling-1) and SOCS-3 in FRTL-5 Thyroid Cells Mol. Endocrinol., March 1, 2000; 14(3): 440 - 448. [Abstract] [Full Text]
|
|
|
|
 |
|
 K. Suzuki, A. Mori, K. J. Ishii, J. Saito, D. S. Singer, D. M. Klinman, P. R. Krause, and L. D. Kohn Activation of target-tissue immune-recognition molecules by double-stranded polynucleotides PNAS, March 2, 1999; 96(5): 2285 - 2290. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
