| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Thyroid Unit Department of Medicine Beth Israel Deaconess Medical Center and Harvard Medical School Boston, Massachusetts 02215
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
NCoR and SMRT are modular proteins (see Fig. 1
) that contain at least three repressing
domains in their N termini, and two domains that mediate interactions
with the TR and RAR isoforms in their C termini (21, 22, 23, 24). In addition
to mediating interactions with mSin3, Sun-CoR (25), and other members
of the corepressor complex, the central domains of NCoR and SMRT also
appear to mediate interactions with the AML-ETO product, which may
prevent normal differentiation and lead to acute myelogenous leukemia
in patients with this chromosomal translocation (26). In addition, the
extreme amino-terminal domain of NCoR appears to be important in the
regulation of the mature protein through its interaction with mSIAH2,
which allows for proteolytic degradation of NCoR (27).
|
The interactions between the corepressors and the TR isoforms are also
influenced by complex formation. We and others have demonstrated that
the TRß1 isoform recruits NCoR preferentially as a homodimer on DNA
and that the addition of retinoid X receptor (RXR) causes a diminution
in corepressor binding. Indeed, NCoR appears to stabilize the homodimer
complex in solution where it normally does not form (24). In contrast,
work using the mammalian two-hybrid system and the TR
1
ligand-binding domain (LBD) has demonstrated that RXR can enhance
interactions with nuclear corepressors (29). However, this study
investigated TR-corepressor interactions in the absence of an
underlying thyroid hormone response element, and did not utilize
full-length TRs. To address these issues, we studied TRß1 and RAR
and examined their ability to recruit either NCoR or SMRT to their
cognate response elements. In addition, we examined the ability of RXR
to influence corepressor recruitment in the context of specific nuclear
receptors. By using a TR mutant that is defective in its ability to
homodimerize, we demonstrate that the TR complex present on a native
response element determines the nature of the corepressor recruited.
Our data demonstrate that the polypeptides that represent the
corepressor interacting domains appear to recognize both specific
nuclear receptors and the complexes that they form.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). We have shown previously using
in vitro translated (IVT) proteins that the TRß1 isoform
prefers NCoR over SMRT (24). Similar data are seen in electrophoretic
mobility shift assay (EMSA) using the GST fusion proteins representing
both interacting domains of NCoR and SMRT on a DR+4 element (Fig. 3A; compare, for example, lanes 3 and 5,
or 7 and 9). Both NCoR and SMRT bind TRß1 on this thyroid hormone
response element (TRE), but NCoR is bound significantly more avidly
than SMRT. This preference is seen over a wide range of amount of GST
protein used, although it is seen most dramatically with lower amounts
of protein (20–100 ng). The shifts associated with corepressor binding
result in a decreased amount of remaining TRß1 homodimer (compare
lane 1 with lanes 3, 7, and 11). The addition of RXR causes a loss of
corepressor binding (lanes 4, 6, 8, 10, 12, and 14). There is no
further supershift in the presence of RXR (see especially lanes 4 and
8), consistent with specific binding to the TR homodimer. These data
suggest that the TRß1 homodimer binds corepressors more avidly than
does the TRß1-RXR heterodimer.
|
|
), which is a retinoic acid
response element, RAR binds corepressor solely when heterodimerized
to RXR (compare lanes 9 and 10, and 13 and 14). Furthermore, the
RAR/RXR heterodimer binds SMRT, but does not interact well with NCoR
(compare, for example, lanes 7–8 with 9–10). The shifts associated
with SMRT binding result in a decreased amount of remaining RAR/RXR
heterodimer (compare lane 2 with lanes 10 and 14). In contrast to
TRß, when 20 ng (lanes 3–6) or 100 ng (data not shown) of GST
protein are used, no specific binding is detected, and it is only at
higher amounts (lanes 7–14) that the preferential binding of the
RAR/RXR heterodimer to SMRT is identified. However, when identical
amounts of protein are used (e.g. 1 µg), the RAR-RXR
heterodimer prefers to interact with SMRT, while the TRß1 homodimer
prefers to interact with NCoR.
To delineate which of the interacting domains (IDs) mediates this
specificity, we next performed EMSA with TRß1 or RAR using
bacterially expressed GST proteins containing individual interacting
domains (Fig. 4, A and B). As Fig. 3 had
demonstrated that differences in corepressor binding to TRß1 are most
apparent with 20 ng of GST fusion protein (lanes 3–6), this amount of
protein was used in Fig. 4A; similar data were obtained when 1
µg of GST fusion proteins was used (data not shown). In contrast,
since binding to RAR could only be clearly detected using higher
amounts of GST fusion proteins, 1 µg of GST protein was used in the
EMSA in Fig. 4B.
|
, the proximal ID 2 allows for specificity
in that N2 strongly binds the TRß1 homodimer (lane 5) while S2 does
not bind the TRß1 isoform well (lane 7) on the DR+4 element. In
contrast, on the retinoic acid response element (RARE) (Fig. 4B), RAR interacts well with S2 (lanes 7–8) and does not bind N2
well (lanes 5–6). These equivalent domains share only 23% homology at
the amino acid level, suggesting that specificity is encoded for within
the ID 2 polypeptide. With the amount of GST protein used in each of
these experiments, neither ID 1 from NCoR or SMRT was able to bind to
either receptor complex on DNA response elements. However, binding of
N1 to the TR homodimer could be seen when greater amounts of protein (1
µg) were used. In addition, N1 binding was enhanced by sequences 3'
to its defined boundary (N1', see Fig. 2), but never approached the
strength of N2 binding. Interestingly, the individual interacting
domains (for example N2 and S2) bound optimally to TR and RAR in the
absence of RXR (compare Fig. 4A, lanes 5 and 6, and Fig. 4B, lanes 7
and 8). Although the RAR homodimer is not believed to exist in
vivo, these data suggest that under certain circumstances it might
be stabilized by interacting proteins. Importantly, however, in the
context of both interacting domains together, RAR binds corepressors
solely as a heterodimer with RXR (Fig. 3B, lanes 10 and 14).
To confirm that indeed the proximal ID 2 was responsible for
receptor specificity, we next swapped the distal ID 1 region among the
NCoR and SMRT constructs, creating chimeric interacting domains
consisting of N2S1 and S2N1 (Fig. 4, C and D). As expected, the TRß1
isoform homodimer preferred the N2S1 chimera (Fig. 4C, lane 3), while
the RAR/RXR heterodimer preferred to interact with the S2N1 chimera
(Fig. 4D, lane 6). Note that since the chimeric constructs bound TR and
RAR more weakly than did wild-type constructs, 2 µg of chimeric
constructs were used in Fig. 4D. However, these data again suggest that
the proximal interacting domains mediate the preferential interactions
of TR and RAR for NCoR and SMRT, respectively, and that that the distal
IDs do not exert restrictive properties in this context. Moreover,
while the individual S2 domain bound RAR well in the absence of RXR
(see above), the addition of either S1 or N1 resulted in enhanced
binding to the RAR-RXR heterodimer (see Fig. 3B, lane 10; and
Fig. 4D, lane 6). These data suggest that portions of the NCoR and SMRT
interacting domains aid in the recognition of receptor complexes
(homodimer or heterodimer).
To define the effects of ligand on the interactions between nuclear
receptors and the individual corepressor interacting domains, we
performed an EMSA with TRß1 in the presence or absence of
T3. As shown in Fig. 5A, TRß1 prefers to bind NCoR over SMRT
(compare lanes 5 and 9). The addition of T3
results in decreased homodimer formation (lanes 1 and 3). In addition,
both NCoR and SMRT are released in the presence of
T3 (compare lanes 5 and 7, and lanes 9 and 11).
As shown in Fig. 5B, TRß1 binds strongly to N2 (lane 1), and this
interaction is disrupted by T3 (lane 3). Finally,
TRß1 binds weakly to S2, and only in the absence of
T3 (lanes 5 and 7).
|
. GST-NCoR includes NCoR amino acids
2063–2300 downstream of GST. IVT-NCoR consists of NCoR amino acids
1579–2454. If the TRß1 homodimer bound two NCoR molecules, we would
expect a mobility shift between the two individual bands produced
(consisting of a TRß1 homodimer binding one GST-NCoR and one
IVT-NCoR). As the TRß1 homodimer forms separate complexes with each
NCoR protein, it appears more likely that the
TRß1 homodimer binds a single NCoR molecule. We next performed a
similar experiment using two different sized N2 constructs: GST-N2
(which contains amino acids 2063–2142) and IVT-N2 (which contains
amino acids 1579–2211). As shown in Fig. 6B, again no intermediate
band is seen, suggesting that each homodimer complex is able to bind a
single interacting domain. This is specific to the TR homodimer, as
Figs. 3B and 4D above suggest that the presence of a second interacting
domain does influence receptor complex recruitment.
|
, the R429Q mutation in context of
the TRß1 isoform is unable to homodimerize well and does not recruit
NCoR either as a homodimer or a heterodimer (lanes 5–6). However, the
R429Q-RXR heterodimer is able to recruit the SMRT interacting domains
to the DR+4 binding site (lane 8). Of note, the R429Q mutant does not
bind either NCoR or SMRT well at low amounts (i.e. 20 ng;
lanes 1–4) and required higher amounts of GST corepressor construct
than did wild-type TRß1 to exhibit strong corepressor binding in
EMSA. However, these data are consistent with the hypothesis that the
structure of the complex influences specific corepressor
recruitment.
|
337T. This mutant TR does not bind
T3, and patients with this mutation have severe
generalized resistance to thyroid hormone (GRTH) (30). In contrast to
R429Q TRß1, the 337T TRß1 mutant homodimerizes well on thyroid
hormone response elements. As shown in Fig. 7B, 337T TRß1 binds
NCoR more strongly than SMRT (compare lanes 1 and 3). As with wild-type
TRß1, this preference is seen most dramatically at lower amounts of
GST fusion proteins used (compare lanes 1–4 with 5–8), although it is
also seen at higher amounts (compare lanes 5 and 7). Thus, 337T
TRß1 is an example of a TR mutant that exhibits enhanced homodimer
formation and preserved corepressor specificity.
SMRT and NCoR Recruit Distinct Nuclear Receptor Complexes in
Cells
To complement the EMSA assay and determine whether SMRT and NCoR
would prefer to interact with homo- or heterodimers, we employed a
two-hybrid assay in mammalian cells. In this system, we fused the
interacting domains of NCoR and SMRT to the Gal4-DNA-binding domain and
fused full-length nuclear receptors downstream of the VP-16 activation
domain. Full-length nuclear hormone receptors were used instead of the
corresponding LBDs, as we and others have demonstrated that the amino
termini influence complex formation (33, 34). In Fig. 8A, the basal activity of Gal4-NCoR or
Gal4-SMRT is set at 1; the interactions between the Gal4-corepressor
constructs and the nuclear receptor-VP16 constructs is then expressed
as relative luciferase activity. As shown in Figure 8B, the experiments
were next done in the presence of cotransfected RXR-VP16 to examine the
effect of heterodimerization on corepressor interactions. These data
are presented as fold expression in the presence vs. absence
of cotransfected RXR-VP16. A ratio greater than 1 implies greater
luciferase activity in the presence of RXR.
|
(upper panel), transfection of
TRß1-VP16 with Gal4-NCoR causes approximately 80-fold stimulation
over the activity of Gal4-NCoR alone. In contrast, transfection of
RAR-VP16 with Gal4-NCoR causes only 25- to 30-fold stimulation over
the activity of Gal4-NCoR. Finally, R429Q TRß1-VP16 only minimally
interacts with Gal4-NCoR in the absence of RXR. We next used the
identical paradigm with Gal4-SMRT as the bait (Fig. 8A, lower
panel). Cotransfection of RAR-VP16 with Gal4-SMRT caused a
40-fold stimulation over the activity of Gal4-SMRT alone. In contrast,
TRß1-VP16 caused only a 6- to 7-fold stimulation in luciferase
activity. Again, R429Q TRß1-VP16 did not interact well in the absence
of cotransfected RXR. All of the nuclear receptor-VP16 constructs
interact to a similar degree with Gal4-RXR (data not shown), suggesting
that they are expressed at similar levels in cells.
These constructs were then cotransfected with RXR-VP16; as shown
in Fig. 8B, cotransfection with RXR-VP16 actually decreased the
interactions between TRß1-VP16 and Gal4-NCoR. In contrast, the
interactions of R429Q TRß1-VP16 with Gal4-NCoR was enhanced 18-fold
by the cotransfection of RXR. Although RXR-VP16 decreased interactions
between RAR-VP16 and Gal4-NCoR (upper panel),
cotransfection of RXR-VP16 enhanced the interactions between this
receptor and Gal4-SMRT (Fig. 8B, lower panel). Additionally,
our previous data (Figure 3B) had showed that, in the presence of a
RARE, RXR is important for the binding of RAR to SMRT. In addition,
although R429Q TRß1-VP16 interacted minimally with Gal4-SMRT in the
absence of RXR, the interaction was enhanced more than 20-fold when
RXR-VP16 was cotransfected. In fact, in the mammalian two-hybrid
system, the interaction between TRß1-VP16 and Gal4-SMRT was also
enhanced by the cotransfection of RXR (
6-fold). Although RXR-VP16
itself interacted minimally with Gal4-SMRT (data not shown), it did not
interact to an extent where it could affect the synergistic level of
interactions seen. Thus, SMRT appears to particularly favor receptor
heterodimers. Moreover, R429Q TRß1, a mutant TR, appears to interact
with corepressors mainly as a heterodimer with RXR, both in solution
and on a thyroid hormone response element.
While RXR-VP16 was used to keep the total amount of VP16 moiety
constant in each dimer pair, we next repeated these experiments in the
presence of RXR-pKCR2. As shown in Fig. 8C, similar results are seen
when RXR-pKCR2 is used instead (although the values are less
pronounced). Again, the presence of RXR is more important in the
context of receptor interactions with SMRT. In addition, cotransfection
of RXR-pKCR2 increased interactions between R429Q TRß1 and the
Gal4-corepressor constructs 5- to 10-fold (data not shown).
While other studies have suggested that RXR can enhance the
interaction between TRß1 and NCoR, these studies used the LBD of the
TR linked to VP16 (29). We therefore used the same paradigm as above in
Fig. 8D, in the context of the nuclear receptor LBDs. In these
experiments, RXR-VP16 had significant impact on the strength of the
interaction of the R429Q LBD (117-fold), and, in contrast to the
full-length TR (see above), also stimulated interactions when
cotransfected with the wild-type LBD (although only 9-fold). In
addition, the ability of Gal4-SMRT to recruit both wild-type and R429Q
mutant LBDs was greatly enhanced by the addition of RXR. In contrast to
the interactions seen with Gal4 NCoR, the interaction of the LBD of
R429Q with Gal4 SMRT was enhanced by more than 300-fold by the addition
of VP16-RXR, consistent with the preferences noted on the DR+4 element
in EMSA. Both the absolute luciferase activity (with RXR) and the
fold-enhancement by RXR were greater in these experiments (as opposed
to experiments using full-length nuclear receptors), suggesting that
the amino termini, and potentially the DNA-binding domains, of TR
modulate interactions with corepressors.
To examine the specificity of the individual interacting domains
in cells, constructs containing N2 or S2 were placed downstream of the
Gal4 DNA binding domain, and used in a similar mammalian two-hybrid
system. As shown in Fig. 8E (upper panel), transfection of
TRß1-VP16 with Gal4-N2 causes approximately 100-fold stimulation. In
contrast, cotransfection with RAR-VP16 results in only about half that
level of activity. While interactions with Gal4-S2 were weaker than
with Gal4-N2, Fig. 8E (lower panel) shows that Gal4-S2
interacts strongly with RAR-VP16, but only minimally with TRß1-VP16,
consistent with the EMSA data (e.g. Fig. 4B).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Recent work has suggested that the L/I-x-x-I/V-I motifs in the corepressor IDs are required for interactions with nuclear receptors (46, 47, 48). Both NCoR and SMRT contain two separate IDs, which have been independently identified by a number of separate groups. The more proximal ID2 of murine and human NCoR shares limited sequence homology with the homologous region of SMRT (23%), whereas the distal ID 1 of murine and human NCoR share approximately 50% amino acid homology. Based on these differences, we and others have hypothesized that the two corepressors may differentially recruit nuclear receptors. Initial studies using EMSA demonstrated that murine (m) NCoR and human (h) SMRT could interact with both the TR and RAR on their respective response elements (3, 4). However, these studies did not examine specificity in context of the interactions, nor did they examine the complexes with which the corepressors may interact. Zamir et al. (49) demonstrated that both TR isoforms could interact with NCoR and SMRT on a thyroid hormone response element (TRE) but did not assess differences in binding, nor did they examine the specificity of the individual interaction domains. Wong and Privalsky examined interactions between individual interaction domains and a number of nuclear receptor isoforms. These data were generated in solution assays (GST pull-down and mammalian two-hybrid assays) and did not incorporate the role of DNA binding into corepressor recruitment. However, these results showed that specificity in corepressor recruitment exists, especially in the context of RAR isoforms and the amino acid sequences present in the hinge region (28). More recently, Hu and Lazar showed that the distal NCoR interacting domain interacts well with RXR, whereas the proximal domain vastly prefers TR (46). These data suggested that the corepressor interacting domains might recognize receptor complexes, which we have examined in this report.
In the present study, we have examined the role of the TRß1 and
RAR isoforms to recruit nuclear corepressors to TREs and RAREs,
respectively. In addition, we have further evaluated the role of RXR in
corepressor recruitment to ascertain the role of homo- vs.
heterodimerization in corepressor recruitment. Our data suggest that
the TRß1 homodimer preferably recruits NCoR while the RAR/RXR
heterodimer preferably recruits SMRT on DNA response elements. This
specificity appears to map the more proximal ID 2 region, as shown in
Fig. 4. Furthermore, chimeric corepressors require the SMRT ID 2 region
(S2N1 or S2S1) to interact with the RAR. In contrast, the ID 2 of NCoR
preferably binds the TRß1 homodimer. These data are consistent with
the marked differences in the ID 2 regions between NCoR and SMRT and
suggest that this region may exert specificity in the recognition of
the TR and RAR.
Given the differences in the ability of the TR and RAR to recruit NCoR
and SMRT, we next investigated the ability of TRß mutants found in
kindreds with RTH to recruit the corepressors to a TRE. Surprisingly,
R429Q TRß1 exhibited altered corepressor specificity in the context
of an underlying TRE. R429Q TRß1 preferred to interact with SMRT
rather than NCoR on a DR+4 element and did not interact with
corepressors well as a homodimer. Previous studies by others using the
R429Q mutant demonstrated an inability to release SMRT as compared with
the wild-type TRß1 isoform but did not examine the preference of the
mutant TR for SMRT (40). Another RTH mutant 337T (30), which forms
strong homodimers, preferably recruits NCoR to the DR+4 element.
The mammalian two-hybrid system also displayed strong differences between the R429Q mutant and wild-type TRß1. In this assay, the interactions between the nuclear receptors and corepressors occur in solution, so the effects of underlying DNA response elements can not be assessed. However, this assay was used to assess the role of heterodimerization in corepressor recruitment in cells. In fact, the addition of RXR enhanced the interaction of R429Q with NCoR and to a greater degree with SMRT, suggesting that its heterodimer form allows for the recruitment of SMRT. In contrast, RXR decreased the interaction of wild-type TRß1 with NCoR, but increased interactions with SMRT in this context. Interactions between RAR and SMRT were also enhanced by the presence of RXR. Thus, SMRT appears to prefer to interact with nuclear receptor-RXR heterodimers. Although SMRT can independently interact with RXR in solution, the absolute value of this interaction is weak when compared with the synergy imparted by the addition of RXR to TR or RAR. Thus, the heterodimer is favorably recognized by SMRT, while homodimer binding recognizes NCoR. Previous work by Zamir et al. (22) demonstrated that the orphan receptor RevErb, which binds DNA only as a homodimer, can recruit only NCoR but not SMRT; these findings are consistent with the data demonstrated here, suggesting that the corepressor interacting domains may recognize complexes as well as specific nuclear receptors.
In addition, the experiments performed with the LBD alone of the wild-type and mutant receptor fused to VP16 demonstrate that the A/B and DNA-binding domains influence corepressor recruitment in the presence or absence of RXR and help explain differences seen by a number of groups in the role of RXR in corepressor recruitment (28, 29). The A/B domain (33, 34) and DNA-binding domain (50) have been shown to influence receptor complex formation on TREs. Alterations in complex formation may explain their influence on TR-corepressor interactions. Alternatively, portions of the A/B and/or DNA-binding domains might conceivably interact directly with the nuclear corepressors.
The data presented here reinforce the need to examine cofactor interactions in the context of DNA binding, given the likely restrictions placed on the receptor once it is bound to DNA. Solution interactions performed by us and others have demonstrated equivalent interactions between NCoR and SMRT and the TR, whereas the introduction of a TRE brings out the differences in the isoforms in their ability to recruit corepressors. Indeed, Wong and Privalsky demonstrated, using solution interactions, that the TR bound equally well to both of the NCoR interaction domains, while on the DR+4 response element it is clear that N2 is preferred over N1. In addition, the S1 domain appears to act to enhance binding to the heterodimer, suggesting that it plays a role in identifying complexes when bound to DNA, consistent with its ability to bind RXR (28, 46).
In summary, the work discussed here demonstrates that both the complex
formed by the nuclear receptor and the DNA response element present in
the responsive promoter can dictate the preference for the corepressor
that is recruited. In addition, the corepressor interacting domains
appear to have preference for specific nuclear receptors, such that the
RAR isoform prefers to interact with SMRT, which is mediated by the
proximal S2 domain. It will be interesting to discern which portion of
the interacting domains are important in the recognition of nuclear
receptors and how complex formation furthers this specificity.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
), which were then placed in
frame downstream of the GST moiety. GST-NCoR includes amino acids (aa)
from human NCoR corresponding to mNCoR, aa 2063–2300; GST-N2
includes aa 2063–2142; and GST-N1 includes aa 2239–2300 (GST-N1'
includes aa 2239–2453). GST-SMRT includes hSMRTe (7), aa
2098–2507; GST-S2 includes aa 2098–2266; and GST-S1 includes aa
2267–2507. The chimeric N2S1 and S2N1 constructs were made by
amplifying the separate domains with an artificial XhoI site
introduced at the 3'-end of the proximal domain and 5'-end of the
distal domain. The pieces were then ligated into PGEX 4T1 as
EcoRI fragments. GST-N2S1 includes amino acids from hNCoR
corresponding to mNCoR, aa 2063–2142, and hSMRTe aa 2267–2507;
GST-S2N1 includes hSMRTe, aa 2098–2266, and NCoR, aa 2143–2300. All
GST constructs were confirmed by dideoxy sequencing.
GAL4-NCoR and GAL4-SMRT were created by ligating the interacting
domains of the respective corepressors downstream of the sequence
encoding the GAL4 DNA-binding domain in the SV40-driven expression
vector pECE. The VP-16 TRß1 and RAR fusions were created by
introducing an in-frame EcoRI site at the 5'-end of the
receptor using PCR and ligating them downstream of the VP16 activation
domain in AASV-VP16. They include aa 1–461 of hTRß1, and aa 1–462
of RAR. The VP-16 R429Q mutant was made by introducing the mutant
sequence as a PstI-Asp718 fragment into the wild-type
construct. The VP16-RXR fusion was created by PCR and consists of aa
2–462 of the human isoform. IVT NCoR (aa 1579–2454) and IVT-N2 (aa
1579–2211) were cloned into pKCR2 and have been described previously
(24, 51). Gal4-N2 was made by placing NCoR, aa 1579–2211, downstream
of the Gal4 DNA-binding domain. Similarly, Gal4-S2 consists of SMRT, aa
2098–2266, downstream of GAL4 DBA binding domain. Construct integrity
was confirmed by restriction endonuclease digestion and dideoxy
sequencing.
GST Fusion Proteins
GST fusion proteins were expressed either in DH5 or BL21
Escherichia coli expressing thioredoxin by induction with
0.1 mM
isopropylthio-ß-D-galactosidase at 30 C (24).
The proteins were isolated by lysis with lysozyme and purified on
Sepharose beads. The bound proteins were eluted using a glutathione
buffer. Verification of protein synthesis was obtained on SDS-PAGE. The
amount of protein generated was quantified using the Bradford
assay.
EMSA
EMSAs were carried out as previously described with either a
32P-radiolabeled DR+4 or DR+5 probe (52).
GST-corepressor fusion proteins (GST-CoRs) were purified on Sepharose
beads and eluted using a glutathione buffer. Nuclear receptors were
in vitro translated in reticulocyte lysate (Promega Corp., Madison WI) using T7 polymerase. IVT NCoR or IVT N2 was
translated similarly. For each EMSA, 4 µl of IVT TR or RAR were used.
For experiments using RXR, 2 µl were used (or an equivalent amount of
unprogrammed reticulocyte lysate as a control). The amount of GST
protein used in each EMSA was identical and is indicated in each
figure. Incubations were carried out for 20 min, and complexes were
resolved on a 5% nondenaturing gel, followed by auto-
radiography.
Cell Culture and Transfection
All transient transfections were performed in CV-1 cells, which
were maintained as described previously (24). Transient transfections
were performed in six-well plates using the calcium phosphate
technique, with each well receiving 1.7 µg of the upstream activity
sequence (UAS-TK) luciferase reporter and 20 ng of a cytomegalovirus
(CMV) ß-galactosidase expression vector. Each well received 80 ng of
Gal4-corepressor construct, along with 80 ng of VP16-TR or -RAR
construct. The addition of 80 ng VP-16 RXR was controlled for by the
presence of empty vector AASV VP-16 (EV-VP16). Similarly, the addition
of 320 ng pKCR2-RXR was controlled with empty vector pKCR2. Fifteen to
18 h after transfection, the cells were washed in PBS and refed
with 10% steroid hormone- depleted FBS. To remove steroid and
thyroid hormones, FBS was treated with 50 mg/ml activated charcoal
(Sigma, St. Louis, MO) and 30 mg/ml anion exchange resin
(type AGX-8, analytical grade, Bio-Rad Laboratories, Inc.
Richmond, CA), as previously described (24). Forty to 44 h after
transfection the cells were lysed and assayed for luciferase and
ß-galactosidase activity. Experiments were performed two to three
times in triplicate. ß-Galactosidase activity was used to control for
transfection efficiency.
The data shown are the pooled results ± SEM and are presented as relative or fold luciferase activity. In particular, the interactions between Gal4-NCoR (or Gal4-SMRT) and nuclear receptor-VP16 constructs are presented relative to luciferase activity in the presence of Gal4-NCoR (or Gal4-SMRT) alone. Fold luciferase activity was measured in the presence vs. absence of cotransfected RXR-VP16 to determine the effect of heterodimerization on interactions between Gal4-NCoR (or Gal4-SMRT) and nuclear receptor-VP16 constructs. For each experiment, the total amount of VP16 construct transfected was held constant with empty vector-VP16 (EV-VP16).
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This work was supported by NIH Grants to R. Cohen (DK-02581), F. Wondisford (DK-53036), and A. Hollenberg (DK-56188 and DK-02354).
Received for publication November 8, 1999. Revision received February 1, 2000. Accepted for publication March 6, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
modulates DNA binding and receptor dimerization. J Biol Chem 273:10270–10278
(RARa) and PZLF-RARa oncoproteins associated with acute promyelocytic
leukemia. Proc Natl Acad Sci USA 94:9028–9033 recruit histone deacetylase in promyelocytic leukemia.
Nature 391:815–818[CrossRef][Medline]
and
TIF1ß in the epigenetic control of transcription by nuclear
receptors. EMBO J 15:6701–6715[Medline]
This article has been cited by other articles:
 |
|
 |
|
  D. S. Machado, A. Sabet, L. A. Santiago, A. R. Sidhaye, M. I. Chiamolera, T. M. Ortiga-Carvalho, and F. E. Wondisford From the Cover: A thyroid hormone receptor mutation that dissociates thyroid hormone regulation of gene expression in vivo PNAS, June 9, 2009; 106(23): 9441 - 9446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Kim, Y. L. Son, and Y. C. Lee Involvement of SMRT Corepressor in Transcriptional Repression by the Vitamin D Receptor Mol. Endocrinol., February 1, 2009; 23(2): 251 - 264. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Astapova, L. J. Lee, C. Morales, S. Tauber, M. Bilban, and A. N. Hollenberg The nuclear corepressor, NCoR, regulates thyroid hormone action in vivo PNAS, December 9, 2008; 105(49): 19544 - 19549. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Hodgson, H. C. Shen, A. N. Hollenberg, and S. P. Balk Structural basis for nuclear receptor corepressor recruitment by antagonist-liganded androgen receptor Mol. Cancer Ther., October 1, 2008; 7(10): 3187 - 3194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Liu, J. Li, P. Shen, B. Husna, E. S. Tai, and E. L. Yong A Natural Polymorphism in Peroxisome Proliferator-Activated Receptor-{alpha} Hinge Region Attenuates Transcription due to Defective Release of Nuclear Receptor Corepressor from Chromatin Mol. Endocrinol., May 1, 2008; 22(5): 1078 - 1092. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Y. Kim, O. G. Park, J. W. Lee, and Y. C. Lee One- plus Two-hybrid System, a Novel Yeast Genetic Selection for Specific Missense Mutations Disrupting Protein/Protein Interactions Mol. Cell. Proteomics, October 1, 2007; 6(10): 1727 - 1740. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. F. R. Velasco, M. Togashi, P. G. Walfish, R. P. Pessanha, F. N. Moura, G. B. Barra, P. Nguyen, R. Rebong, C. Yuan, L. A. Simeoni, et al. Thyroid Hormone Response Element Organization Dictates the Composition of Active Receptor J. Biol. Chem., April 27, 2007; 282(17): 12458 - 12466. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Moriyama, T. Tagami, T. Usui, M. Naruse, T. Nambu, Y. Hataya, N. Kanamoto, Y.-s. Li, A. Yasoda, H. Arai, et al. Antithyroid Drugs Inhibit Thyroid Hormone Receptor-Mediated Transcription J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 1066 - 1072. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Takeshita, K. Inagaki, J. Igarashi-Migitaka, Y. Ozawa, and N. Koibuchi The endocrine disrupting chemical, diethylhexyl phthalate, activates MDR1 gene expression in human colon cancer LS174T cells. J. Endocrinol., September 1, 2006; 190(3): 897 - 902. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Lefebvre, C. Brand, S. Flajollet, and P. Lefebvre Down-Regulation of the Tumor Suppressor Gene Retinoic Acid Receptor {beta}2 through the Phosphoinositide 3-Kinase/Akt Signaling Pathway Mol. Endocrinol., September 1, 2006; 20(9): 2109 - 2121. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Y. Wu, R. N. Cohen, E. Simsek, D. A. Senses, N. E. Yar, H. Grasberger, J. Noel, S. Refetoff, and R. E. Weiss A Novel Thyroid Hormone Receptor-{beta} Mutation That Fails to Bind Nuclear Receptor Corepressor in a Patient as an Apparent Cause of Severe, Predominantly Pituitary Resistance to Thyroid Hormone J. Clin. Endocrinol. Metab., May 1, 2006; 91(5): 1887 - 1895. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-H. Huang, C.-Y. Lee, P.-J. Tai, C.-C. Yen, C.-Y. Liao, W.-J. Chen, C.-J. Liao, W.-L. Cheng, R.-N. Chen, S.-M. Wu, et al. Indirect Regulation of Human Dehydroepiandrosterone Sulfotransferase Family 1A Member 2 by Thyroid Hormones Endocrinology, May 1, 2006; 147(5): 2481 - 2489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. R. Johnson, C.-W. Li, L.-Y. Chen, J. C. Ghosh, and J. D. Chen Regulation and Binding of Pregnane X Receptor by Nuclear Receptor Corepressor Silencing Mediator of Retinoid and Thyroid Hormone Receptors (SMRT) Mol. Pharmacol., January 1, 2006; 69(1): 99 - 108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Mani, H. Huang, S. Sundarababu, W. Liu, G. Kalpana, A. B. Smith, and S. B. Horwitz Activation of the Steroid and Xenobiotic Receptor (Human Pregnane X Receptor) by Nontaxane Microtubule-Stabilizing Agents Clin. Cancer Res., September 1, 2005; 11(17): 6359 - 6369. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Meng, P. Webb, Y.-F. Yang, M. Shuen, A. F. Yousef, J. D. Baxter, J. S. Mymryk, and P. G. Walfish E1A and a nuclear receptor corepressor splice variant (N-CoRI) are thyroid hormone receptor coactivators that bind in the corepressor mode PNAS, May 3, 2005; 102(18): 6267 - 6272. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Yu, K. Markan, K. A. Temple, D. Deplewski, M. J. Brady, and R. N. Cohen The Nuclear Receptor Corepressors NCoR and SMRT Decrease Peroxisome Proliferator-activated Receptor {gamma} Transcriptional Activity and Repress 3T3-L1 Adipogenesis J. Biol. Chem., April 8, 2005; 280(14): 13600 - 13605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Lee and M. L. Privalsky Heterodimers of Retinoic Acid Receptors and Thyroid Hormone Receptors Display Unique Combinatorial Regulatory Properties Mol. Endocrinol., April 1, 2005; 19(4): 863 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Goodson, B. A. Jonas, and M. L. Privalsky Alternative mRNA Splicing of SMRT Creates Functional Diversity by Generating Corepressor Isoforms with Different Affinities for Different Nuclear Receptors J. Biol. Chem., March 4, 2005; 280(9): 7493 - 7503. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Hodgson, I. Astapova, S. Cheng, L. J. Lee, M. C. Verhoeven, E. Choi, S. P. Balk, and A. N. Hollenberg The Androgen Receptor Recruits Nuclear Receptor CoRepressor (N-CoR) in the Presence of Mifepristone via Its N and C Termini Revealing a Novel Molecular Mechanism for Androgen Receptor Antagonists J. Biol. Chem., February 25, 2005; 280(8): 6511 - 6519. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Li and J. Y. L. Chiang Mechanism of rifampicin and pregnane X receptor inhibition of human cholesterol 7{alpha}-hydroxylase gene transcription Am J Physiol Gastrointest Liver Physiol, January 1, 2005; 288(1): G74 - G84. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C M Klinge, S C Jernigan, K A Mattingly, K E Risinger, and J Zhang Estrogen response element-dependent regulation of transcriptional activation of estrogen receptors {alpha} and {beta} by coactivators and corepressors J. Mol. Endocrinol., October 1, 2004; 33(2): 387 - 410. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Villa, J. Santiago, B. Belandia, and A. Pascual A Response Unit in the First Exon of the {beta}-Amyloid Precursor Protein Gene Containing Thyroid Hormone Receptor and Sp1 Binding Sites Mediates Negative Regulation by 3,5,3'-Triiodothyronine Mol. Endocrinol., April 1, 2004; 18(4): 863 - 873. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. I. Castillo, R. Sanchez-Martinez, J. L. Moreno, O. A. Martinez-Iglesias, D. Palacios, and A. Aranda A Permissive Retinoid X Receptor/Thyroid Hormone Receptor Heterodimer Allows Stimulation of Prolactin Gene Transcription by Thyroid Hormone and 9-cis-Retinoic Acid Mol. Cell. Biol., January 15, 2004; 24(2): 502 - 513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. De Bosscher, W. Vanden Berghe, and G. Haegeman The Interplay between the Glucocorticoid Receptor and Nuclear Factor-{kappa}B or Activator Protein-1: Molecular Mechanisms for Gene Repression Endocr. Rev., August 1, 2003; 24(4): 488 - 522. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Hu, S. Li, J. Wu, C. Xia, and D. S. Lala Liver X Receptors Interact with Corepressors to Regulate Gene Expression Mol. Endocrinol., June 1, 2003; 17(6): 1019 - 1026. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Townson, K. M. Dobrzycka, A. V. Lee, M. Air, W. Deng, K. Kang, S. Jiang, N. Kioka, K. Michaelis, and S. Oesterreich SAFB2, a New Scaffold Attachment Factor Homolog and Estrogen Receptor Corepressor J. Biol. Chem., May 23, 2003; 278(22): 20059 - 20068. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Liang, P. Webb, A. Marimuthu, S. Zhang, and D. G. Gardner Triiodothyronine Increases Brain Natriuretic Peptide (BNP) Gene Transcription and Amplifies Endothelin-dependent BNP Gene Transcription and Hypertrophy in Neonatal Rat Ventricular Myocytes J. Biol. Chem., April 18, 2003; 278(17): 15073 - 15083. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Makowski, S. Brzostek, R. N. Cohen, and A. N. Hollenberg Determination of Nuclear Receptor Corepressor Interactions with the Thyroid Hormone Receptor Mol. Endocrinol., February 1, 2003; 17(2): 273 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-Y. Liu and G. A. Brent A Complex Deoxyribonucleic Acid Response Element in the Rat Ca2+/Calmodulin-Dependent Protein Kinase IV Gene 5'-Flanking Region Mediates Thyroid Hormone Induction and Chicken Ovalbumin Upstream Promoter Transcription Factor 1 Repression Mol. Endocrinol., November 1, 2002; 16(11): 2439 - 2451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Takeshita, M. Taguchi, N. Koibuchi, and Y. Ozawa Putative Role of the Orphan Nuclear Receptor SXR (Steroid and Xenobiotic Receptor) in the Mechanism of CYP3A4 Inhibition by Xenobiotics J. Biol. Chem., August 30, 2002; 277(36): 32453 - 32458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Cheng, S. Brzostek, S. R. Lee, A. N. Hollenberg, and S. P. Balk Inhibition of the Dihydrotestosterone-Activated Androgen Receptor by Nuclear Receptor Corepressor Mol. Endocrinol., July 1, 2002; 16(7): 1492 - 1501. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Jepsen and M. G. Rosenfeld Biological roles and mechanistic actions of co-repressor complexes J. Cell Sci., February 15, 2002; 115(4): 689 - 698. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Marimuthu, W. Feng, T. Tagami, H. Nguyen, J. L. Jameson, R. J. Fletterick, J. D. Baxter, and B. L. West TR Surfaces and Conformations Required to Bind Nuclear Receptor Corepressor Mol. Endocrinol., February 1, 2002; 16(2): 271 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. N. Cohen, S. Brzostek, B. Kim, M. Chorev, F. E. Wondisford, and A. N. Hollenberg The Specificity of Interactions between Nuclear Hormone Receptors and Corepressors Is Mediated by Distinct Amino Acid Sequences within the Interacting Domains Mol. Endocrinol., July 1, 2001; 15(7): 1049 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Hu, Y. Li, and M. A. Lazar Determinants of CoRNR-Dependent Repression Complex Assembly on Nuclear Hormone Receptors Mol. Cell. Biol., March 1, 2001; 21(5): 1747 - 1758. [Abstract] [Full Text]
|
|
|
|
|
|
K. Hashimoto, F. H. Curty, P. P. Borges, C. E. Lee, E. D. Abel, J. K. Elmquist, R. N. Cohen, and F. E. Wondisford An unliganded thyroid hormone receptor causes severe neurological dysfunction PNAS, March 1, 2001; (2001) 51454698. [Abstract] [Full Text]
|
|
|
|
|
|
K.-h. Lin and Y.-h. Wu shen-liang chen Impaired Interaction of Mutant Thyroid Hormone Receptors Associated with Human Hepatocellular Carcinoma with Transcriptional Coregulators Endocrinology, February 1, 2001; 142(2): 653 - 662. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Yoh and M. L. Privalsky Transcriptional Repression by Thyroid Hormone Receptors. A ROLE FOR RECEPTOR HOMODIMERS IN THE RECRUITMENT OF SMRT COREPRESSOR J. Biol. Chem., May 11, 2001; 276(20): 16857 - 16867. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hashimoto, F. H. Curty, P. P. Borges, C. E. Lee, E. D. Abel, J. K. Elmquist, R. N. Cohen, and F. E. Wondisford An unliganded thyroid hormone receptor causes severe neurological dysfunction PNAS, March 27, 2001; 98(7): 3998 - 4003. [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 |
