| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(Received for publication, September 1, 1994; and in revised form, November 28, 1994) From the
Thyroid hormone receptors (TRs) bind specific thyroid hormone
response elements (TREs) as heterodimers with retinoid X receptors
(RXRs) and act as transcriptional activators. As homodimers, TRs can
bind a distinct set of sequences and function as ligand sensitive
repressors. In our study, we compared the natural malic enzyme TRE
(ME-TRE) as a model system for the TR/RXR heterodimer pathway to the
chicken lysozyme silencer element F2-TRE which is strongly bound and
regulated by TR/TR homodimers. Using electrophoretic mobility shift
assays, transient transfections with a variety of natural and synthetic
triiodothyronine and thyroxine derivatives as well as limited
proteolytic analysis, we show that the natural homo- and heterodimeric
pathways show similar ligand requirements. Furthermore, we observe that
the ligand-induced conformational changes in the receptor proteins that
either result in a loss of TR/TR homodimer binding and release of
transcriptional repression or in transcriptional activation of TR/RXR
heterodimers are indistinguishable. Therefore, we propose that in TR/TR
homodimers and TR/RXR heterodimers very similar moieties of the
receptors are involved in ligand binding and subsequent conformational
changes that lead to loss of gene repression (TR/TR homodimer) and gain
of gene activation (TR/RXR heterodimer). Thyroid hormones regulate important biological processes such as
metamorphosis, development, growth, homeostasis and general metabolism (1) . The major forms of thyroid hormones comprise
3,5,3`,5`-tetraiodo-L-thyronine (L-T4) ( More recent evidence suggests that TRs can also form heterodimers
with other receptors(30, 31) . In addition TR/TR
homodimers (33, 34) ( The major goal of this study
was to compare these two major T3 signaling response pathways, the
TR/RXR heterodimeric and the TR/TR homodimeric pathway, in terms of
ligand mediated action. For this we analyzed the T3 induced release of
repression by TR/TR homodimers using the F2-TRE, comparing it to TR/RXR
heterodimers induced transcriptional activation on the ME-TRE,
employing natural and synthetic thyroid hormone analogs. We show here
that the TR/TR homodimeric and the TR/RXR heterodimeric pathways have
very similar ligand requirements. Our data support the idea that the
ligand-induced conformational changes in TR that lead to
transcriptional activation by TR/RXR heterodimers and release of
repression by TR/TR homodimers are identical or at least very similar.
This implies that the orientation and spacing of the half-sites in the
various natural TREs is a key determinant for either transcriptional
activation or release of repression by TRs.
Figure 1:
Two natural TREs have distinct receptor
binding properties. A, sequences of wild type TREs used in
this analysis. The hexameric repeats involved in receptor binding and
function as demonstrated by previous studies are indicated by arrows.
The rat malic enzyme TRE (ME-TRE) sequence from 288 to 259 (45) and the chicken lysozyme silencer F2-TRE sequence from
-2354 to -2326 (14) are shown. B, binding
of homo- and heterodimeric complexes.
Figure 2:
Analysis of TR/TR homodimer and TR/RXR
heterodimer response pathways in vivo. A, ligand
dependent transactivation of the ME TRE in CV-1 cells. The TR/RXR
heterodimeric complex was analyzed on a natural DR-4 element, the rat
malic enzyme (ME) TRE, in transient transfection experiments.
The ME-TRE was cloned upstream of the tk-promoter generating
the reporter construct pBLCAT2 ME TRE. 100 ng of pBLCAT2 ME TRE
together with 5 ng of expression vectors were cotransfected into CV-1
cells. The cells were treated with different concentrations of hormone (L-T3) and 24 h later tested for CAT activity. The mean of
four experiments is shown. The standard deviations are indicated. B, ligand dependent release of repression by TR/TR-homodimer.
The chicken lysozyme silencer F2-TRE, which has an inverted palindromic
structure with a spacing of 6 nucleotides, was cloned downstream the
TATA-box into the pBLCAT2 vector (pBLCAT2 TATA F2). This
construct has constitutive promoter activity which is repressed in the
presence of TR. T3 releases repression and results in the induction of
the reporter gene. 300 ng of pBLCAT2 TATA F2 were cotransfected into
CV-1 cells either with 50 ng TR
When we analyzed the F2-TRE cloned upstream
of a tk-CAT gene, cotransfection of TR
Figure 3:
Dose response curves of natural and
synthetic T3 and T4 analogs. Repression release and transactivation
potencies of the different thyromimetics are shown. TR
Figure 4:
Activities of T3 analogs. The EC
We analyzed
[
Figure 5:
Ligand-induced conformational changes
analyzed by limited proteolytic digestion. A, homodimer
analysis.
We were also interested on how
the presence of different response elements could influence the
conformation of the thyroid hormone receptor in the TR/TR homodimeric
(F2-TRE) and TR/RXR heterodimeric (ME-TRE) pathways. Using the limited
chymotrypsin digestion assay, the liganded TRs appeared to have the
same structural configurations when interacting as a TR/TR homodimer
with the F2-TRE (Fig. 5A) or as a TR/RXR heterodimer
with the ME-TRE (Fig. 5B). Furthermore, TR/RXR
heterodimer binding to the inverted palindrome F2-TRE or the
palindromic TREp induced very similar structural changes in TR as
judged from the limited proteolytic analyses. The unliganded TR yielded
analogous digestion patterns in the presence of DNA as in its absence.
These results suggest that the inverted palindromic F2-TRE, the DR-4
element ME-TRE and the palindromic TREp(9) , although different
in their architecture, do not induce diverse structural states in TRs
when analyzed in this system. With the isolation and characterization of an increasing
number of TREs, different T3 signaling mediated by TRs have become
apparent(44, 53) Our analysis in addition suggests that the observed
thyromimetic potencies of the T3 analogs reflect the binding affinities
to the TRs. From studies of T3 and T4 analogs in quantitative
structure-activity analysis (54, 55) it is known that
the important 3`-residue interaction is strictly lipophilic and limited
in size. The 4`-hydroxyl is suggested to donate a hydrogen bond to the
ligand binding pocket of the TR. Consequently, when a hydrophilic group
is introduced in 3`-position (compound 312810) or when the 4`-hydroxyl
is blocked by a methyl group (compound 322010) the thyromimetic
potencies should be eliminated, as observed here (Fig. 3).
Interestingly, a lipophilic ring substitution of the Iodine at the 3`
position (compound 322085) increased the biological activity (Fig. 5B). In contrast, introduction of a flexible and
bulky phenyl group that increased the size of the 3` substituent
(compounds 322450, 322094, and 312693) led to a considerable drop in
biological activity. Changing the 3`-residue position to the
2`-position (compound 322010) led to a further reduction of the
activity. Taken collectively, our experimental data confirm the known
qualitative structure-binding affinity relationship for the interaction
of L-T3 with TRs present in nuclear extracts (54, 55, 56) . The subtle differences
observed between TR Recently, it
has been shown that L-T3 binding to TRs induces conformational changes
in the receptor, as measured by electrophoretic mobility analysis of a
receptor-DNA complex(33, 61, 62) , circular
dichroism (CD) spectroscopy(63) , or limited protease digestion
assays (64, 65) . Even though gel retardation analysis
showed enhanced migration of the monomeric and dimeric forms of the TRs
in the presence of ligand, these effects were rather small compared to
ligand induced disruption of TR/TR homodimers. While the gel
retardation technique is most frequently used to study ligand induced
effects, the CD spectroscopy method is a more effective analysis to
explore ligand induced secondary structure changes in TRs as
demonstrated for the chicken TR by Toney et al.(63) .
Interestingly, these investigators observed in addition to the ligand
effect also a measurable change in the CD spectrum when adding the
palindromic TREp, but no effect was detected in the presence of a DR-4.
These data could indicate that the TREp induces an additional
conformational change in the chicken TR In
conclusion, the results presented here suggest that the ligand-induced
changes in the TRs required for activation of the homodimeric and the
heterodimeric response pathways are very similar. Thus the ligand
binding pocket of TRs does not appear to be in very different
configurations when TRs are complexed as heterodimers or homodimers on
different structural elements. However, some structural changes, not
detectable with the tools used here that might also contribute to
differential TR function cannot be excluded. TRs are for instance also
known to interact with the transcription factor AP-1 (66) .
Recent results indicate (67) that this nuclear receptor/AP-1
interaction may require different ligand-induced conformational changes
than transcriptional activation. Whether TR/RXR, TR/RAR(32) ,
TR/VDR (30) , and TR/PPAR (31) heterodimeric complexes
have similar ligand responsiveness still needs to be analyzed.
Volume 270,
Number 7,
Issue of February 17, 1995 pp. 3107-3114
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
)and
3,3`,5-triiodo-L-thyronine (L-T3), the latter is the most
active form. The biological effects of these hormones are mediated by
specific nuclear thyroid hormone receptors (TRs). Two genes encoding
two different receptor subtypes, TR
and TR
, have been
characterized (see for review, see (2, 3, 4, 5) ). The receptors are
members of the steroid hormone/retinoic acid receptor superfamily, a
large group of transcription
factors(6, 7, 8) . TRs have dual regulatory
roles and can function as transcriptional activators as well as
transcriptional repressors(5, 9) . Like other members
of the superfamily, TRs mediate T3 signals through specific DNA
sequences, the T3 response elements (TREs), usually found in the
promoter regions of responsive genes. The characterization of TREs
revealed that several configurations of two half-sites of the sequence
AGGTCA (or derivatives of this sequence) are possible, including a
palindrome with no spacer (10) , direct repeats with 4 base
pair (bp) spacer(11, 12, 13) , and inverted
palindromes/everted repeats spaced by 6 bp(14, 15) .
For effective DNA interaction, the TRs were found to require
association with a nuclear auxiliary factor, called
TRAP(16, 17, 18, 19, 20) ,
now identified as retinoid X receptors
(RXRs)(21, 22, 23, 24, 25, 26) .
The RXRs also play a central role in several other signal transduction
pathways since they heterodimerize with several other nuclear
receptors, including the retinoic acid receptors (RARs), the vitamin D
receptor (VDR), and the peroxisome proliferator activated receptors
(PPARs)(21, 22, 23, 24, 25, 26, 27, 28, 29) . 
)and TR monomers (35, 36) have been proposed to confer T3 responsive
transcriptional regulation increasing the variety of possibilities for
T3 signal transduction mechanisms. However, so far it appears that most
TREs are activated by the TR/RXR heterodimeric pathway since the
heterodimers have the highest affinity for these TREs. TR/RXR
heterodimers were shown to form in solution in the absence of
ligand(21, 22, 25) , while the presence of
specific DNA binding sites like the -myosin heavy chain TRE or the
malic enzyme TRE (ME-TRE) were shown to strongly enhance this
dimerization(37) . Interestingly, the chicken lysozyme TRE
(F2-TRE), an IP-6 type response element, initially characterized as a
silencer (14) can be bound by TR/TR homodimers and TR/RXR
heterodimers, the TR/TR homodimers forming the more stable complexes
with this TRE. TR/TR homodimer binding to the F2-TRE,
however, is inhibited by T3(38) allowing the
homodimers to function as T3 sensitive
repressors(5, 16) .
Reagents
Restriction enzymes, T4-ligase, and
polynucleotide kinase were from New England Biolabs, Stratagene, and
Boehringer Mannheim. Radiolabeled L-[
S]methionine (1175 Ci/mmol) and
-[
P]dATP (3000 Ci/mmol) were purchased from
DuPont-NEN. 
C-Labeled low range protein molecular markers
were obtained from Life Technologies, Inc. Poly(dI-dC), thyroid
hormones (L-T3 and L-T4), and thyroid hormone analogs
(3,3`,5-triiodo-D-thyronine (D-T3);
3,5,3`,5`-tetraiodo-D-thyronine (D-T4);
3,3`,5`-triiodo-L-thyronine (rT3); 3,3`,5-triiodothyroacetic
acid (TRIAC)) were purchased from Sigma. The L-T3 analogs (322085,
322450, 322094, 312693, 320989, 320883, 322010, 312810) were kindly
provided by Dr. Brian L. Buckwalter, American Cyanamid Company,
Princeton, NJ.Recombinant Plasmids
The reporter construct
pBLCAT2 ME carried a single copy of the sequence
CAGGACGTTGGGGTTAGGGGAGGACAGTGG (recombinant ME from -288 to
-259) in the BamHI site of pBLCAT2(39) . The
pBLCAT2 TATA reporter vector was constructed by creating a BglII site 3` adjacent to the TATA box using polymerase chain
reaction and suitable primers. By inserting the F2 sequence
TTATTGACCCCAGCTGAGGTCAAGTTACG (14) into the BglII site
of pBLCAT2 TATA, the pBLCAT2 TATA F2 was generated. The expression
vectors pECE TR and TR have been previously
described(20, 40) .Transient Transfections and Electrophoretic Mobility
Shift Assays
CV-1 cells were used for transient transfection
assays as described(41, 42) . Receptor proteins were
produced by in vitro transcription/translation as described (41, 43) and used in the electrophoretic mobility
shift assay as reported
previously(9, 20, 24, 41) . When
L-T3 was used, receptor proteins were first preincubated with ligand
for 10 min prior adding the labeled DNA. The double-stranded
oligonucleotides F2-TRE (gatccTGACCCCAGCTGAGGTCAg) and ME-TRE
(gatccAGGACGTTGGGGTTAGGGGAGGACAGTGg) had both BamHI overhangs,
used for fill in labeling with the Klenow fragment of DNA polymerase.Limited Proteolytic Digest Analysis
Receptor
proteins were preincubated with the natural and synthetic T3 analogs.
Prior to incubation, the ligands were diluted to 10
M in a buffer containing 50 mM KCl, 1 mM dithiothreitol, 2.5 mM MgCl, 5% glycerol, and
20 mM HEPES at pH 7.91. The final concentration of the
thyromimetics was 10
M. After 10 min at
room temperature 1-µl aliquots containing different amounts of
chymotrypsin were added. Proteolytic digestions were carried out for 10
min and stopped by adding 5 µl of 2 
SDS sample buffer and
boiling for 5 min. The SDS-polyacrylamide gel electrophoresis was done
at room temperature using a 10% polyacrylamide gel
(acrylamide/N,N`-methylene-bisacrylamide, weight
ratio of 29:1). Before vacuum drying at 80 °C, the gels were fixed
in 10% acetic acid and 20% ethanol, followed by a brief incubation in
Amplify solution (Amersham Corp.).
The ME-TRE and the F2-TRE Differ in Their Structure and
Function
Among the numerous natural TREs identified (reviewed in (5) and (44) ) the rat ME-TRE (45) and the
F2-TRE (14) represent two examples of distinct structural
arrangements of the half-sites. The ME-TRE (Fig. 1A)
consist of two direct repeats separated by 4 nucleotides (DR-4). This
type of TRE belongs to a group of elements that also includes the
-myosin heavy chain TRE (46, 47) ,
MoMLV-TRE(25, 48) , mouse myogenin TRE(49) ,
skeletal -actin TRE(50) , rat sarcoplasmic reticulum
Ca
ATPase TRE(51) , and mouse
MyoD-TRE(52) . All of these DR-4 TREs require RXR for TR
binding. In contrast, in the F2-TRE (Fig. 1A) the
half-sites are arranged as an inverted palindrome separated by 6 bp
(IP-6). This IP-6 motif allows high affinity TR/TR homodimer binding as
well as TR/RXR heterodimer binding, when using in vitro synthesized receptors (Fig. 1B). As shown
previously (38) only the TR/TR homodimer binding
to F2-TRE can be disrupted by T3 (Fig. 1B). T3 has no
effect on the affinity of the TR/RXR heterodimer to the ME-TRE or
F2-TRE. When transiently transfected into CV-1 cells, TR allowed
induction of the pBLCAT2 ME-TRE reporter plasmid, in response to
increasing amounts of T3 (Fig. 2A). The fold
stimulation at saturating ligand concentrations varied between 3- and
5-fold under these conditions. Cotransfection of TR together with
the RXR expression plasmid enhanced the CAT expression only
slightly suggesting that CV-1 cells contain sufficient endogenous RXR
to allow DR-4 activation. Indeed nuclear extracts of CV-1 cells
preincubated with in vitro translated TR shifted the
ME-TRE to a position corresponding to the in vitro produced
TR/RXR heterodimers (20 and data not shown), supporting the in vivo data.
P-Labeled ME-TRE and
F2-TRE were incubated with equal amounts of in vitro translated receptors. Binding of TR homodimers in the
presence and absence of 10M L-T3 and
TR/RXR heterodimers were compared. Unprogrammed lysate (Lysate) served as a control.
expression vector or empty vector
as a control and 200 ng of -gal expression vector. The cells were
treated with different concentrations of hormone (L-T3) and 24 h later
analyzed for reporter gene (CAT) activity. CAT values were
normalized to -gal expression. The mean of four experiments is
shown. The standard deviations are indicated. The basal tk-promoter activity was repressed by TRs in the absence of
ligand by more than 80%.
decreased the
basal CAT-activity. Addition of T3 not only reversed this effect, but
also up-regulated the CAT expression severalfold over the basal level,
consistent with previous observations (14) (data not shown). To
be able to separate the heterodimer response from the homodimer release
of repression response, we cloned the F2-TRE downstream of the TATA-box (Fig. 2B) such that the TRE-bound receptors would
inhibit the transcriptional machinery. Only the TR/TR homodimers could
interfere with the transcription initiation in a ligand responsive
manner. Indeed this experimental design allowed to separate the
predominantly TR/TR homodimeric effects from the TR/RXR heterodimeric
effects in transient transfection assays (Fig. 2B).
Repression of the F2-TRE was released in a ligand sensitive way,
reestablishing the basal expression without superactivation. Thus, the
TATA F2 construct is suitable for functional analysis of the release of
transcriptional repression important in the homodimer pathway.Ligand Requirements for Thyroid Hormone Receptors in
Homo- and Heterodimeric Pathways
Having established a system
that allows to measure the ligand response of TR/TR homodimers in
vivo, we compared the ligand-sensitive release of repression with
transcriptional activation by heterodimers. Different T3 and T4 analogs
were compared for their ability to effect gene regulation by homo- and
heterodimers. One important question we wanted to test was whether
natural thyromimetics could act as selective compounds distinguishing
between the two pathways. CV-1 cells were cotransfected with TR in
a transient transfection assay and increasing amounts of L-T3, D-T3,
L-T4, D-T4, TRIAC, and rT3 were applied (Fig. 3A).
Comparison of the sigmoidal derepression and transactivation curves
revealed that the compounds behaved very similar in the release of
repression on the F2-TRE and in the transactivation of the ME-TRE. The
diverse ligand induced effects were highly comparable for the specific
ligands (Fig. 3A). When we included TR into our
analysis (Fig. 3B), we also observed almost identical
dose response curves for the homo- and heterodimers. These results
indicate that TR and TR have highly similar T3 ligand
requirements in both pathways.
(A) and TR (B) were analyzed. Transfections were
carried out essentially as in Fig. 2. CAT activity was measured
after 24 h of incubation in presence of different concentrations of the
T3 and T4 analogs. The values were normalized to -gal expression.
High and medium activity compounds were compared to L-T3 (upper and middle panels) while low activity compounds (lower panels) were compared to
L-T4.
Synthetic Thyromimetics Behave Very Similarly in Homo-
and Heterodimeric Response Pathways
We extended our analysis
with a series of synthetic T3 analogs. Most of the T3 analogs used (Fig. 4B) had substitutions in important positions
(3`-, 3-, 5- substituents). In compounds 322010, 320883, 312693,
322094, and 322450, the iodines were exchanged with bromide in the 3-
and 5-position. Furthermore, these compounds had in the 3`-position a
bulky phenyl group. In compound 320989, the 3` substituent was move to
the 2`-position. Whereas compound 322085 had both 3`- and 2`-positions
involved in a planar naphthalene ring system. In addition to the
lipophilic 3` substitutions (compounds 322085, 322450, 322094, 312693,
320883, and 322010), compound 312810 had a hydrophilic ethyl amino
acetate substituent. First experiments showed that none of these
synthetic thyromimetics could distinguish between TR and TR.
However, their potency to activate gene expression and release gene
repression in CV-1 cells differed markedly for the analogs analyzed (Fig. 3). According to their half maximal effective
concentrations (EC
) measured, four groups of T3 analogs
could be identified (Fig. 4): highly potent, intermediate, weak,
and inactive analogs. The compound 322085, L-T3, and TRIAC were highly
active. The EC of these ligands ranged from 40 pM for compound 322085 to 2.5 nM for TRIAC. Thus 322085 was
up to 3.5-fold more effective than L-T3 in both the transactivation as
well as in the release of repression analysis. D-T3, compound 322450,
compound 322010, and compound 312693 belong to the intermediate group,
having EC values between 7.5 and 150 nM. The weak
group consisted of L-T4, D-T4, and compounds 320989 and 320883, with
EC values of 0.1 µM and lower. The compounds
322010 and 312810 and reverse T3 showed no or only marginal activity (Fig. 3). Disruption of TR/TR homodimer binding to the F2-TRE by
the different T3 analogs was also according to the measured EC values when we used these compounds in the gel retardation assay
(data not shown). Only subtle differences of approximately 2-fold
between TR and TR activation in the homo- and heterodimeric
pathways could be detected (Fig. 4). Overall the thyromimetic
activities of the ligands were essentially the same for the ME-TRE and
F2-TRE. Thus the ligand sensitivity of TRs is not differentially
affected in homodimeric and heterodimeric complexes, indicating that
the TR ligand binding domain is in a similar configuration and equally
accessible in hetero- and homodimers bound to distinct response
elements.
values for the different ligands as determined by transcriptional
activation of the ME-TRE and release of repression on the F2-TRE are
listed. The structural formulas of the natural ligands (A) and
the synthetic T3 analogs (B) are shown. n.a., not
active.
Limited Proteolytic Digestion Analysis of Ligand-induced
Conformational Changes
To further examine whether the TR
proteins have similar ligand-induced structural states in the presence
of different TREs, we examined the hormone- induced structural changes
of TR using a partial proteolytic analysis, where different
proteolytic digestion patterns induced by ligand binding can indicate
structural changes in the analyzed protein.S]methionine-labeled TR incubated with
different concentrations of chymotrypsin. As demonstrated in Fig. 5A, preincubation of labeled TR with T3 gave
a different protease digestion pattern than unbound receptor. The
unliganded TR was more accessible to the protease; digestion with
chymotrypsin at a concentration of 10 µg/ml generated only two
fragments of low molecular weight, whereas the T3-induced conformation
protected the liganded TR from the protease activity. Chymotrypsin at a
concentration of 10 µg/ml digested the liganded TR in a distinct
pattern, yielding more and higher molecular weight fragments (Fig. 5A). In the presence of T3 a protected 30-kDa
peptide fragment appeared when a higher concentration of chymotrypsin
(10 µg/ml) was used. The protease resistant 30-kDa fragment was not
observed in the absence of hormone. The diverse proteolytic digestion
patterns and the T3-induced protection of a 30-kDa fragment reflect the
ligand induced conformational changes of the thyroid hormone receptor.
From immunocoprecipitation studies, it was previously shown that TR/RXR
heterodimers are formed in solution(37) . The protease
digestion pattern of the thyroid hormone receptor, when present as a
heterodimer with RXR in solution (Fig. 5B), showed a
slightly delayed digestion kinetic (a weak high molecular weight band
is still present at 10 µg/ml of chymotrypsin). Whereas the overall
proteolytic digestion pattern was identical to the unliganded thyroid
hormone receptor (Fig. 5B). Addition of T3 to the
TR/RXR heterodimer induced the same proteolytic digestion pattern as
observed for the T3 occupied TR. This delayed digestion kinetic might
be due to the interference of RXR with the digestion, but does not
suggest a major conformational alteration of TR since the protease
digestion pattern remained unchanged.
S-Labeled TR was analyzed by partial
chymotryptic digestion (2 µg/ml, 5 µg/ml, and 10 µg/ml).
Addition of 10M L-T3 yielded a different
partial digestion pattern of TR than in absence of ligand.
Liganded TR was less accessible to the protease. Additionally, a
protected 30 kDa peptide fragment (triangle) was observed that
was not present in absence of ligand. The presence of 1 µM of F2-TRE double stranded oligo did not alter the ligand-induced
protection pattern. B, heterodimer analysis using limited
protease digestion. S-Labeled TR was preincubated
with equal amounts of RXR and with either 1
buffer(-), F2-TRE (F2), TREp (TREp), or ME-TRE (ME) prior to limited chymotryptic digestion. The presence of
RXR and/or the various TREs did not alter the digestion
pattern.
In our study, we have
selected two mechanistically different TR response pathways to analyze
the molecular influence of the ligand. To facilitate this analysis we
have designed a new reporter construct that allows to follow the ligand
sensitive release of repression by TR/TR homodimers on the F2-TRE.
Interestingly, transcriptional activation of the TR/RXR heterodimer and
the release of transcriptional repression mediated by the TR/TR
homodimers appeared to be highly similar in terms of their ligand
requirements. Not only did the heterodimers and homodimers show very
similar T3 response curves, but they also reacted very similarly to T3
and T4 analogs. and TR activation indicate that the
existing panel of T3 analogs does not yet allow sufficient molecular
discrimination to induce/repress TR subtypes selectively. This
situation may be comparable to the retinoid field several years ago,
when no satisfactory receptor selective ligands were available. The
recent discovery of selective retinoids for numerous functions (57, 58, 59, 60) could be followed
by a similar development in the thyromimetics field.. Other laboratories (64, 65) using partial tryptic digestions showed that
the T3 induced conformational changes generated trypsin-resistant
peptide fragments of the ligand binding domain. Using this same
method, we did not observe that DNA binding enhanced or decreased the
T3 induced protected chymotrypsin peptide fragments significantly.
Chymotrypsin produced a protected fragment (30 kDa) in the presence of
T3 that did not appear in absence of ligand, suggesting major
conformational differences between liganded and unliganded receptors.
Additionally, we also observed a T3 induced delayed chymotryptic
TR degradation as previously demonstrated with trypsin, implying
that the conformational changes produced a more compact ligand binding
domain, less accessible, to the proteases. Importantly, our results
showed that the major conformational changes induced by L-T3 are at
least very similar in both heterodimers and homodimers.
))
We thank Dr. G. Salbert and Dr. G. Graupner for
helpful discussions and comments on the manuscript. Additionally, we
thank Dr. B. L. Buckwalter, American Cyanamid Company, for providing us
with T3 analogs, and Shelby Bates for help with preparation of the
manuscript.
1 thyroid hormone receptor induced by binding of 3, 3`,5-triiodo-L-thyronine. Biochem. Biophys. Res. Commun. 195, 385-392
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Ikeda, E. C. Wilcox, and W. W. Chin Different DNA Elements Can Modulate the Conformation of Thyroid Hormone Receptor Heterodimer and Its Transcriptional Activity J. Biol. Chem., September 20, 1996; 271(38): 23096 - 23104. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-O. Lee, M. I. Dawson, N. Picard, P. D. Hobbs, and M. Pfahl A Novel Class of Retinoid Antagonists and Their Mechanism of Action J. Biol. Chem., May 17, 1996; 271(20): 11897 - 11903. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Yen, Y. Liu, A. Sugawara, and W. W. Chin Vitamin D Receptors Repress Basal Transcription and Exert Dominant Negative Activity on Triiodothyronine-mediated Transcriptional Activity J. Biol. Chem., May 3, 1996; 271(18): 10910 - 10916. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |
