| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Instituto de Investigaciones Biomédicas. CSIC-UAM, Madrid, Spain
1Correspondence: Instituto de Investigaciones Biomédicas, Arturo Duperier 4, 29029 Madrid, Spain. E-mail: aaranda{at}iib.uam.es
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: retinoid receptor • half-site • heterodimer • retinoid X receptor
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. These receptors play an important role in diverse biological
processes by stimulating or repressing target gene expression, and
profound changes in the pattern of gene expression could result from an
alteration in the combination and/or concentration of the different
receptors. Nuclear receptors primarily act through binding to specific
DNA sequences known as hormone response elements—vitamin D response
elements (VDREs) in the case of VDR and retinoic acid response elements
(RAREs) in the case of RAR. VDREs and RAREs typically consist of at
least two copies of the consensus sequence
PuGG/TTCA, which can be
configured into a variety of motifs (2)
. The arrangement and spacing
between the motifs are determinant to confer selectivity and
specificity. Data from several laboratories have shown that the most
potent VDREs and RAREs are configured as direct repeats (DRs). Thus, a
DR separated by three nucleotides (DR3) acts as a VDRE (3)
, whereas DRs
separated by two or five nucleotides (DR2 and DR5, respectively) act as
RAREs (4)
. In addition to spacing, small differences in the half-site
sequence and the sequence of the flanking extension of the response
elements also appear to be important parameters in determining receptor
binding efficiency (5)
. Furthermore, widely spaced DRs can act as
promiscuous response elements for different nuclear receptors (6)
. As
opposed to receptors for steroid hormones that bind to DNA as
homodimers, VDR and RAR preferentially bind DNA as heterodimers with
RXR. The formation of heterodimers not only increases the efficiency
with which they bind to DNA, but also results in specific response
element repertoires (7)
. Since DR elements are inherently asymmetric,
heterodimers may bind to them with two distinct polarities. Indeed, it
has now been established that on DR3 and DR5, RXR occupies the 5'
half-site and the heterodimeric partner (e.g., VDR and RAR,
respectively) occupies the 3' half-site (8
9
10
11)
. This particular
arrangement permits RAR and VDR, on ligand binding, to undergo
conformational changes that facilitate the recruitment of a coactivator
complex. This multisubunit complex contains factors with histone
acetylase activity, which allows chromatin remodeling and creates a
permissive state for promoter activation (12
13
14)
. In addition to
ligand-dependent gene activation, some receptors (including RAR) act as
constitutive repressors in the absence of ligand. The transcriptional
corepressors N-CoR (nuclear corepressor) and SMRT (silencing mediator
for retinoic acid and thyroid hormone receptors) interact with the
unliganded receptors (15,
16)
, as well as with mSin3A and histone
deacetylase 1, and function as adaptors to convey a repressive signal
to the transcriptional apparatus (17,
18)
. Ligand binding results in
dissociation of the corepressor complex and the subsequent binding of
coactivators (14,
19)
.
Although binding to the HRE is a prerequisite for transcriptional
stimulation, only a subset of DNA sequences that act as high-affinity
receptor binding elements function as response elements. Thus, although
RXR/RAR heterodimers have been reported to bind a DR separated by one
nucleotide (DR1) with high affinity, in most contexts they are unable
to activate transcription in response to either RAR or RXR ligands
(20)
. As a result, RAR inhibits RXR-dependent transcription from these
sites. The RAR/RXR heterodimer has been reported to bind DR1 elements
with the RAR in the 5' position (20)
. In this reversed polarity, N-CoR
remains associated with the RAR/RXR heterodimer even in the presence of
RAR ligands, resulting in constitutive repression (21)
. These
observations suggest that the response elements can allosterically
regulate interactions with corepressors to determine positive or
negative regulation of gene expression. We have previously observed
that the heterodimer RXR/VDR can bind to RAREs in a transcriptionally
inactive form, and under these circumstances VDR can inhibit the
RAR-dependent stimulation. For instance, in pituitary cells that
coexpress RAR, VDR, and RXR, incubation with vitamin D represses the
response of the growth hormone and RARß2 promoters to RA (22,
23)
.
This repression is mediated by a palindromic RARE that also acts as a
thyroid hormone response element in the case of the growth hormone
promoter and by a DR5-type RARE (ßRARE) in the case of the RARß2
promoter. In this study we have compared binding of RXR/RAR and RXR/VDR
heterodimers to the ßRARE. The RXR/VDR heterodimer binds with high
affinity and in a stable manner to the ßRARE, but it does not
stimulate transcription of a reporter gene containing this element.
This lack of response can be attributed at least in part to the fact
that RXR/VDR does not show a preferred orientation on the DR5 element,
whereas it binds with a normal polarity on the DR3-type VDRE present in
the osteopontin promoter. However, the VDR does not interact with
corepressors either in solution or on the RARE; therefore, the lack of
activation is not due to corepressor binding. Furthermore, the RXR/VDR
heterodimer can recruit coactivators both in the DR3 and DR5 elements.
These results show for the first time that the vitamin D receptor
heterodimer can bind without a defined polarity to a DNA binding
element in a transcriptionally inactive fashion and that a mechanism
different from the inability to dislodge corepressors might be involved
in the activation defect.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
25
26)
between positions -53/-37
(5'-GTCGAGGGTAGGGTTCACCGAAAGTTCACTCGTCGAC-3') in front of the
thymidine kinase promoter of pBL-CAT8+. DR4-tk-CAT and DR3-tk-CAT
contain the same sequence in which one or two oligonucleotides,
respectively, between both hemi-sites of the element have been deleted
(7)
. Expression vectors for human VDR, RXR
, and RAR (26,
27)
cloned in pSG5 have been described previously. pGST-VDR, which
expresses a fusion protein between glutathione S-transferase (GST) and
VDR, was obtained by polymerase chain reaction by using the pSG5-VDR
plasmid as a template and the oligonucleotides
5'-CGGGATCCATGGAGGCAATGGCGG-3' and 5'-GGAATTCTCAGGAGATCTCATTGC-3'
to generate the complete VDR cDNA. This fragment was then subcloned
into BamHI/EcoRI sites of the pGEX-2T plasmid.
Similarly, in GST-RAR, the cDNA of the human RAR was amplified
with the oligonucleotides
5'-GGAATTCCTATGCTGGGTGGACTC-3' and 5'-GCTCTAGATCACGGGGAGTGGGT-3'
and subcloned into pGEX-2T. In GST-
ABRXR, the amino-terminal
109 amino acids containing region A/B of RXR have been deleted. To
obtain thisplasmid, a fragment of 1078 base pairs was
amplified by polymerase chain reaction from pSG-hRXR with the
oligonucleotides 5'-GGAATTCTGATGGGCCTCAATGGCGTCC-3' and
5'-GCTCTAGACTAAGTCATTTGGTGCCG-3'. The fragment obtained was also cloned
into pGEX-2T. The plasmids GST-ACTR (coactivator of thyroid hormone and
retinoic acid receptors) (12)
and GST-SMRT (15)
containing the nuclear
receptor-interacting sequences of both proteins have been described
previously.
Cell culture and transfections
GH4C1 cells were cultured in Dulbecco's modified Eagle's medium
containing 10% fetal calf serum and transfected by electroporation, as
described previously (22,
23)
. The reporter plasmids were mixed with 20
to 30 million GH4C1 cells and exposed to a high-voltage pulse (200–250
V, 960 µF). The cells from each electroporation were split into
different culture plates in medium containing 10% AG1 x 8 resin
charcoal-stripped newborn calf serum. Treatments were
administered in serum free medium. CAT activity was determined by
incubation of the cell extracts with
[14C]-chloramphenicol. The unreacted and
acetylated [14C]-chloramphenicol were separated
by thin-layer chromatography and quantified with an InstantImager. The
data are expressed as the percentage of acetylated forms after each
treatment. Each treatment with the ligands was performed at least in
duplicate cultures, which normally exhibited less than 10% variation
in CAT activity, and the experiments were repeated at least three
times. The results are expressed as the mean ± standard deviation
of the CAT values obtained.
Protein preparations
VDR, RAR, and RXR cloned in pSG5 were used for
in vitro transcription and translation,
following the manufacturer's recommendations for the TNT7 Quick
coupled transcription/translation System (Promega, Madison, Wis.).
Reactions were performed in the presence of 40 µCi of
[35]S-methionine (Amersham, Little Chalfont,
U.K.) (for the pull-down assays) or with the same amount of unlabeled
amino acid (for gel retardation assays). Five microliters of the
reaction product were resolved in 10% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The gel was
dried and autoradiographed. The GST fusion proteins VDR, RAR, ABRXR,
ACTR, or SMRT were expressed in the bacterial strain BL21 (DE3). They
were grown at 37°C in 2 XYT (tryptone 16 g/l, yeast extract, NaCl 5
g/l, pH 7) until the absorbance reached 0.6. Then the induction was
performed at 30°C for 3 h with 0.4 mM isopropyl
b-D-thiogalactopyranoside. GST and GST fusion proteins were
expressed and purified by standard techniques, following the
recommendations of Pharmacia (Piscataway, N.J.). The expression of
correctly sized proteins was monitored by SDS-PAGE.
Mobility shift assays
Gel retardation assays were performed with the
in vitro translated receptors or with
GST-fusion proteins and the DR3 or DR5 oligonucleotides: 5'-ACAA
GGTTCACGAGGTTCACGTCT-3' and 5'-GGGTAGGGTTCACCGAAAGTTCACTCG-3'. For the
binding reaction, the proteins were incubated on ice for 15 min in a
buffer (20 mM Tris HCL (pH7.5), 75 mM KCl, 1 mM dithiothreitol, 5
µg/ml bovine serum albumin, 13% glycerol), containing 3 µg poly
(dI-dC) and then for 15–20 min at room temperature with ~50,000 cpm
of labeled double-stranded oligonucleotide end-labeled with
[32P]dCTP, using Klenow fragment as kinase.
DNA–protein complexes were resolved on 6% polyacrylamide gels in
0.5 x TBE buffer. The gels were then dried and autoradiographed at
-70°C. Representative retardation assays, which were repeated at
least three times with similar results, are shown in the figures.
Dissociation kinetics were studied by gel retardation `off-rate'
curves with in vitro translated RXR/VDR and RAR/VDR
heterodimers. Complexes were performed for 15 min at room temperature
and a 500-fold excess of unlabeled oligonucleotide was added at 0 min
The reactions were applied in a running gel at different time
intervals.
Cross-linking
Gapped oligonucleotides containing BrdU residues were used
for cross-linking experiments (10)
. The DR5 probe consisted of a 5'
half-site, 5'-TCGAGGGTAGGG(BrdU)(BrdU)CACCG-3', and a 3' half-site,
5'-AAAG(BrdU)(BrdU)CACTCGCACTCG-3'. The DR3 probe used
contained the 5' hemi-site 5'-CAGACCAACAAGG(BrdU)(BrdU)CAC-3', and the
3' hemi-site contained 5'GAGG(BrdU)(BrdU)CACGTCTCTAAAGG-3'. These
DR5 and DR3 probes, which contain the RARE of the RARß2 promoter and
the VDRE of the osteopontin promoter, respectively, were annealed to
the corresponding contiguous complementary strand. Either the 5' or 3'
half-sites were labeled with [
-32P]ATP.
Full-length GST-VDR or GST-RAR (80 ng) and/or the same amount of
GST-ABRXR were incubated for 15 min at room temperature with
~500.000 c.p.m. of the gapped probes in a binding buffer containing
10 mM Tri-HCl (pH 8.0), 0.1 mM EDTA, 50 mM dithiothreitol, 80 mM KCl, 2
µg poly (dI-dC), and 5% glycerol. UV cross-linking (using a
wavelength of 312 nm) was performed at 0°C for 30 min. The samples
were boiled for 5 min in a buffer containing SDS; the DNA–protein
complexes were separated in 10% SDS-PAGE, identified by
autoradiography, and when appropriate, quantified by densitometry.
Protein–protein interactions
GST-pull-down assays were performed with 5 µl of the in
vitro translated L-[35S]methionine-labeled
proteins. These proteins were incubated with 1 µg of the GST-fusion
protein or with the same amount of GST as a control, immobilized in
glutathione-Sepharose beads. The proteins were first incubated in the
presence of 100 nM vitamin D, 1 µM RA, or ethanol for 20 min at room
temperature in glass tubes. The reaction with the beads was performed
for 1 h at 4°C in a binding buffer containing 25 mM HEPES KOH,
pH 7.9, 1% glycerol, 5 mM Mg2Cl, 1 mM DTT, 0.05% Triton X-100, 5 mM
EDTA, and 1 mM PMSF. Free proteins were washed from the beads with a
buffer containing increasing concentrations (50, 100, and 200 mM) of
KCl, and the bound proteins were analyzed by SDS-PAGE and
autoradiography.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
shows that incubation with RA resulted in a strong activation of the
DR5 reporter, whereas vitamin D was unable to stimulate this construct.
We have previously shown that vitamin D blocks the response of the
RARß2 promoter to RA (23)
and, consistent with those results, vitamin
D markedly reduced RA inducibility of DR5-tk-CAT. Deletion of one
nucleotide resulted in a substantial loss of response to RA, but did
not increase the response to vitamin D. In contrast, deletion of two
nucleotides converted the RARE into a strong VDRE, although a
significant response of DR3-tk-CAT to RA was still observed. Vitamin D
and RA did not have antagonistic effects, but cooperated to stimulate
the activity of the DR3-containing promoter.
|
Binding of vitamin D and retinoid receptors to DR5 and DR3 elements
As shown in Fig. 2
A, the lack of inducibility of DR5-tk-CAT by vitamin D is not
due to inability of VDR to bind the HRE. Confirming our previous
observations (23)
, the inactive RXR/VDR heterodimer (lane 4) binds to
the DR5 element. We observed in different experiments that binding of
this heterodimer was as strong as that caused by the active RXR/RAR
heterodimer, shown in lane 2. Furthermore, as illustrated in Fig. 2B
, RXR/VDR bound with similar strength to the DR3 (lane 6)
and to the DR5 elements. RXR/RAR also bound to the DR3, although with
less potency than to the DR5. Therefore, the affinity of RAR/RXR
binding to both elements correlated with the activation observed in the
transfection experiments, although there was a clear discrepancy
between the strong binding of RXR/VDR to a DR5 element and the lack of
transcriptional activation by vitamin D.
|
A decreased stability of inactive RXR/VDR complexes vs. active RXR/RAR
complexes on the DR5 element could account for the inability of vitamin
D to stimulate transcription from this element. The stability of these
complexes was assessed by off-rate experiments in which heterodimeric
complexes formed with 32P-labeled DR5 were
challenged with an excess of unlabeled oligonucleotide and loaded at
different times onto a running gel. As shown in Fig. 3
, binding of both RXR/VDR (right panel) and RXR/RAR (right panel) to the
DR5 was rather stable on challenge. Quantitation of results obtained in
different experiments showed that active and inactive complexes
disappeared with the same kinetics and that ~70% of the label
remained even after 30 min of incubation. Therefore, this result
indicates that the DR5 is a strong binding element for the
transcriptionally inactive RXR/VDR heterodimer.
|
RXR/VDR shows no preferred polarity on the DR5 element
Since DR elements are asymmetric, heterodimers may bind to them
with two distinct polarities (e.g., 5'-RXR, 3'-VDR, or 5'-VDR, 3'-RXR).
To analyze the polarity of binding to the DR5 and DR3 elements,
recombinant GST receptors were cross-linked with UV light to
BrdU-substituted DRs that contained a gap between the two hemi-sites.
Either the 5' or the 3' oligonucleotide of the gapped strand was
labeled with 32P and annealed to a continuous
complementary strand. The sizes of VDR and RAR complexes with the
radiolabeled probe are rather similar and run with a similar mobility
in denaturing SDS-PAGE. To distinguish the cross-linked products in the
heterodimers, a shorter amino-terminally truncated RXR (ABRXR) was
used. Lanes 1 to 6 in Fig. 4
shows binding of VDR, RAR and RXR to the DR5. Although, as illustrated
in Fig. 2
, homodimers bind the element with low affinity, a detectable
homodimeric binding under the conditions used in the cross-linking
experiments was found. Cross-linking showed that these receptors
bind indistinctly to the 5' or 3' hemi-sites. When ABRXR/RAR
heterodimers were cross-linked, the rapidly migrating ABRXR bound
covalently to the 5' motif (lane 9), whereas RAR (which run slowly)
always bound to the 3' motif (lane 10). No such binding polarity of
ABRXR/VDR heterodimers was observed. As illustrated in lanes 7 and
8, VDR bound to both the 5' and the 3' hemi-sites with a similar
intensity, and the same was true for ABRXR. Quantification of bands
from three independent experiments showed that 56 ± 5% of VDR
was bound to the upstream motif and 44 ± 5% to the downstream
motif. In contrast, the ABRXR/VDRR heterodimer showed a defined
polarity on a DR3 element, with ABRXR occupying the upstream motif
and VDR the downstream hemi-site.
|
The inactive RXR/VDR heterodimer can recruit coactivators on a DR5
element
Ligand-dependent transcriptional activation by nuclear receptors
requires binding of coactivators to the AF-2 region. It was possible
that the conformation acquired by the RXR/VDR heterodimer on the DR5
element would prevent coactivator binding to the VDR. To test this
possibility, electrophoretic mobility shift experiments with DR5 and
DR3 elements in the presence of receptors and the coactivator ACTR were
conducted. As shown in Fig. 5
(left panel), ACTR does not bind the DR5 element (lanes 14 to 16), but
incubation of RXR/VDR (lane 10) or RXR/RAR (lane 5) with the
coactivator causes the formation of RXR/VDR/ACTR and RXR/RAR/ACTR
complexes detectable as supershifts (lanes 12 and 7, respectively). In
the absence of ACTR, incubation with ligands only causes a slight
increase in the mobility of the receptor heterodimers (lanes 6 and 11).
However, addition of vitamin D (lane 13) or RA (lane 8) greatly
enhances association of the receptor–DNA complexes with ACTR; under
these conditions, the heterodimers are totally supershifted. Similar
results, excepting the fact that the RAR/RXR heterodimer bound weakly
to the element, were obtained with a DR3 (Fig. 5
, left panel).
Therefore, the recruitment of ACTR to RXR/VDR-DNA complexes was
identical on the DR5 (where the heterodimer is transcriptionally
inactive) and a DR3 (which acts as a potent response element). This
suggests that altered coactivator binding does not account for the
differential activities of VDR on the two elements.
|
RXR/VDR does not bind the corepressor SMRT on the DR5
It had been shown that on DR1 elements, nuclear corepressors
remain associated with RXR/RAR heterodimers even in the presence of
ligand, resulting in constitutive repression (21)
. As shown in
Fig. 6
A, in the absence of DNA, VDR does not appear to interact
with the corepressor SMRT. A fusion protein of GST with the silencing
mediator of retinoid and thyroid hormone receptors (SMRT) was unable to
pull-down 35S-labeled, in
vitro-translated VDR. In contrast, GST-SMRT associated strongly
with 35S-RAR in the absence of ligand, and RA
reduced this interaction by more than 75%. Despite the finding that
VDR does not appear to bind SMRT in the absence of DNA, we next
investigated the possibility that corepressors might interact
differentially with RXR/VDR heterodimers in a DNA-specific fashion. For
this purpose, gel retardation experiments similar to those shown in
Fig. 5
, but using SMRT, were performed. The left panel in Fig. 6B
illustrates representative results obtained with the DR5,
and the right panel shows the results with the DR3 element. Incubation
of RXR/RAR heterodimers (lane 5 in both panels) with SMRT eliminated
the RXR/RAR-DNA complex and resulted in a supershifted complex (lanes
7), indicating that SMRT was binding to the heterodimer on DNA.
Addition of RA releases SMRT from the complexes (lanes 8). In contrast,
incubation with SMRT did not cause a supershift of the RXR/VDR
heterodimer on the DR5 or the DR3 in the absence or presence of vitamin
D (lanes 10 to 13). This shows that corepressor binding does not
account for the inability of VDR to activate transcription from the
DR5.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
proposed a simple rule to determine the
response element specificity of the different heterodimeric complexes.
According to this rule, RXR/VDR heterodimers should bind to a DR3,
whereas RAR/RXR heterodimers should bind to a DR5. Later results have
shown that recognition of DRs by different receptor heterodimers is
more complex (2,
4
5
6
7)
. Our results also indicate some promiscuity in
binding, since VDR/RXR can bind not only to a DR3 but also to a DR5
element composed of two half-sites of the motif -PuGTTCA. It has been
proposed that the DNA binding domains (DBDs) of several receptors are
responsible for the selection of the binding elements (5
6
7
8
9
10,
28)
;
regions within the VDR DBD that confer selectivity for direct repeats
with appropriate spacing and direct asymmetrical protein–protein
contacts have been identified (29)
. Our data suggest that the DBD of
VDR should be capable of extensive rotation and bending with respect to
the carboxyl-terminal ligand binding domains (LBDs) to allow
heterodimeric binding to elements spaced by five and three nucleotides.
Alternatively, different heterodimerization interfaces may be used in
RXR/VDR binding to DR5 and to DR3 motifs.
In the presence of its cognate ligand, VDR/RXR heterodimers
trans-activate from a DR3 and not from a DR5, although they
bind these elements with similar efficiency. Therefore, there is a
marked discrepancy between the strong in vitro binding of
RXR/VDR to the DR5 and the lack of transcriptional activation in
vivo. This situation is reminiscent of that found with the RAR/RXR
heterodimer bound to a DR1. On DR1 elements, RXR/RAR heterodimers
exhibit no response to activating ligands and repress RXR-dependent
transcription (20)
. We have also observed that vitamin D represses the
response of the DR5-containing promoter RARß2 to RA (23)
, and our
present results show that the same regulation is conferred by this DR5
to a heterologous promoter. A similar repressive effect of vitamin D on
RA signaling on DR2-type RAREs has also been observed (30)
. The
different transcriptional efficiency of the RXR/RXR heterodimers on DR1
and DR5 elements appears to originate from the different polarities of
the heterodimeric complexes. On DR5 sites, RAR/RXR heterodimers bind in
such a way that RAR occupies the downstream half-site, whereas on DR1
elements, RAR occupies the upstream half-site (8
9
10)
. A similar
polarity of RXR/VDR heterodimers on DR3 elements has been described
previously (11,
31)
. Our cross-linking experiments confirm that DR3
complexes display the 5' RXR/VDR polarity. In contrast to the defined
polarity on a DR3, we also demonstrate that DR5 complexes are composed
of an equal mixture of 5' RXR/VDR and 5' VDR/RXR orientations. Rotation
of the DBDs with respect to the LBDs must occur in order to allow VDR
and RXR to switch between the upstream and downstream half-sites of the
DR5 element. The fact that binding occurs in both orientations on a DR5
again suggests that RXR and VDR interactions could generate two
distinct dimerization interfaces or one very flexible dimerization
interface.
The finding that RXR/VDR shows no preferred polarity on a DR5 was
unexpected, as anisotropic binding of RXR/VDR heterodimers occurs not
only on DR3 motifs, but also on natural VDREs formed by imperfect
inverted palindromes (32,
33)
. Although unusual, heterodimeric binding
to a DR element without a defined polarity has been described
previously. Similar to our results with RXR/VDR on a DR5, no binding
polarity of RXR/RAR heterodimers on a widely spaced DR10 motif is
observed, even though this heterodimer binds cooperatively to such
elements (6)
. In contrast, the DR10 motif confers responsiveness to RA.
These findings, as well as the complete rather than partial lack of
response to vitamin D, suggest that other mechanisms besides the
absence of polarity could contribute to the incapacity of the RXR/VDR
heterodimer to stimulate transcription from a DR5. For example, the
conformation of the DBDs of RXR/VDR bound to DR5 and DR3 elements may
be different, which may facilitate or exclude interaction with other
regulatory factors. It has been shown that binding of coactivators to
the AF-2 domain is required for ligand-dependent
trans-activation by the RXR/VDR heterodimer (34,
35)
. This
raised the possibility that binding to a DR5 could prevent recruitment
of coactivators to the AF-2 region. Our in vitro data show
that the coactivator ACTR associates equally well, and in a vitamin
D-dependent manner, to RXR/VDR bound to DR5 and DR3 elements. This
result excludes the possibility that the ligand binding ability of VDR
on a DR5 was abolished. However, the fact that in vivo
recruitment of coactivators occurs on a DR5 remains to be established.
On the other hand, the molecular explanation for repression by
RAR on a DR1 element reflects the failure of ligand to cause the
dissociation of corepressors rather than altering coactivator
association (21)
. However, this model cannot be applied to repression
by VDR on a DR5. Our results show that VDR did not bind the corepressor
SMRT in solution in a manner that was detectable. Nevertheless, it has
been described that corepressor function is affected by steric effects
related to DNA binding in a receptor-specific manner. Thus,
heterodimerization with RXR provides the interaction surface necessary
for corepressor recruitment by other receptors on DNA (36,
37)
. It was
then possible that the RXR/VDR heterodimer could bind corepressors on
the DR5. However, our results clearly demonstrate that SMRT does not
associate with RXR/VDR bound to either a DR5 or a DR3. The possibility
that unidentified corepressors different from SMRT could be involved in
the lack of transcriptional activation via a DR5 cannot be dismissed,
but since in the absence of ligand we have been unable to detect a
repressive effect of VDR in transfection assays in vivo,
this possibility is highly unlikely.
Therefore, although an altered polarity that could prevent corepressor release provided an attractive hypothesis for the lack of activation of RXR/VDR on a DR5, our results suggest a greater complexity. It is likely that direct or indirect interactions between the bound receptor heterodimers and other neighboring transcription factors or components of the basal transcriptional machinery might be influenced by the conformation of the heterodimers in the differently spaced DRs. In addition, the requirement for unidentified additional factors that could be essential for trans-activation by RXR/VDR heterodimers and could recognize specifically the DR3-bound complex, but not the DR5-bound complex, cannot be ruled out. In any case, binding of RXR/VDR heterodimers to DRs with different transcriptional outcomes may generate selectivity and expand the interactions between these receptors and other factors to create a sensitive and specific transcriptional complex.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication October 19, 1998.
Revision received January 21, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. I. Castillo, R. Sanchez-Martinez, A. M. Jimenez-Lara, A. Steinmeyer, U. Zugel, and A. Aranda Characterization of Vitamin D Receptor Ligands with Cell-Specific and Dissociated Activity Mol. Endocrinol., December 1, 2006; 20(12): 3093 - 3104. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Voss, I. A. Demarco, C. F. Booker, and R. N. Day Functional interactions with Pit-1 reorganize co-repressor complexes in the living cell nucleus J. Cell Sci., August 1, 2005; 118(15): 3277 - 3288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Scsucova, D. Palacios, M. Savignac, B. Mellstrom, J. R. Naranjo, and A. Aranda The repressor DREAM acts as a transcriptional activator on Vitamin D and retinoic acid response elements Nucleic Acids Res., April 22, 2005; 33(7): 2269 - 2279. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-N. Bastie, N. Balitrand, F. Guidez, I. Guillemot, J. Larghero, C. Calabresse, C. Chomienne, and L. Delva 1{alpha},25-Dihydroxyvitamin D3 Transrepresses Retinoic Acid Transcriptional Activity via Vitamin D Receptor in Myeloid Cells Mol. Endocrinol., November 1, 2004; 18(11): 2685 - 2699. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Bi and P. K. Rogan Bipartite pattern discovery by entropy minimization-based multiple local alignment Nucleic Acids Res., September 23, 2004; 32(17): 4979 - 4991. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. K. Keeton, T. M. Fletcher, C. T. Baumann, G. L. Hager, and C. L. Smith Glucocorticoid Receptor Domain Requirements for Chromatin Remodeling and Transcriptional Activation of the Mouse Mammary Tumor Virus Promoter in Different Nucleoprotein Contexts J. Biol. Chem., July 26, 2002; 277(31): 28247 - 28255. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Aranda and A. Pascual Nuclear Hormone Receptors and Gene Expression Physiol Rev, July 1, 2001; 81(3): 1269 - 1304. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mathur, P. W. Tucker, and H. H. Samuels PSF Is a Novel Corepressor That Mediates Its Effect through Sin3A and the DNA Binding Domain of Nuclear Hormone Receptors Mol. Cell. Biol., April 1, 2001; 21(7): 2298 - 2311. [Abstract] [Full Text]
|
|
|
|
|
|
R. M. Tolón, A. I. Castillo, A. M. Jiménez-Lara, and A. Aranda Association with Ets-1 Causes Ligand- and AF2-Independent Activation of Nuclear Receptors Mol. Cell. Biol., December 1, 2000; 20(23): 8793 - 8802. [Abstract] [Full Text]
|
|
|
|
|
|
G. Perez-Juste, S. Garcia-Silva, and A. Aranda An Element in the Region Responsible for Premature Termination of Transcription Mediates Repression of c-myc Gene Expression by Thyroid Hormone in Neuroblastoma Cells J. Biol. Chem., January 14, 2000; 275(2): 1307 - 1314. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
