| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Metabolic Research Unit (A.M., W.F., H.N., J.D.B., B.L.W.) and Departments of Biochemistry and Biophysics and Cellular and Molecular Pharmacology (R.J.F.), University of California San Francisco, San Francisco, California 94143; Center for Endocrinology, Metabolism, and Molecular Medicine (J.L.J.), Northwestern University Medical School, Chicago, Illinois 60611; and Clinical Research Institute, Center for Endocrine and Metabolic Diseases (T.T.), Kyoto National Hospital, Kyoto 612-8555, Japan
Address all correspondence and requests for reprints to: Dr. Brian L. West, Molecular Biology Department, Plexxikon, Inc., 91 Bolivar Street, Berkeley, California 94710. E-mail: bwest{at}plexxikon.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Corepressors like nuclear receptor copressor (N-CoR) (4), also called RIP13 (5), and silencing mediator of retinoid and thyroid receptor (SMRT) (6), also called TRAC (T3 receptor associating cofactor) (7), were identified by their binding to the TR and RAR, but binding of varying strength has been detected to a number of different nuclear receptors including the RXR (5), VDR (8), PPAR (9, 10), ER (11, 12), PR (13, 14, 15), and the orphan receptors RevErb (16, 17), chicken ovalbumin upstream promoter transcription factor 1 (18), DAX-1 (19), steroidogenic factor-1 (20), and RVR (17). To understand corepressor function it is important to determine how they interact with receptors. Central to this is to define site(s) that bind these proteins. Initially it was proposed that the corepressor binds to a CoR box, which includes helix 1 of the TR LBD based on the observation that deletion of this helix abolished N-CoR binding to the TR (4). This notion received further support by the observations that combined mutation of three conserved residues (A228, H229, and T232 AHT) within the CoR box region and of a residue preceding helix 1 (P214 in the hTRß) blocked binding of N-CoR and SMRT (4, 6, 7). However, the AHT and P214R mutations are of residues that are buried in the liganded TR structure (21), and these mutations and deletion of helix 1 may affect corepressor binding through conformational changes near to or distant from the helix 1 region.
Alterations of nuclear receptors outside helix 1 have also been reported to diminish binding of corepressors. In the TR these include deletion of the helix 11 (22) and mutations located in residues of helices 3, 4, and 5 that are surface exposed in the ligand-bound structure (23, 24). Evidence for a role of helices 3 and 5 has also been reported for the orphan receptors RevErb and RVR (17) and helix 11 in RevErb (16). Hormone binding stabilizes particular conformations of helix 12 relative to helices 3 and 5 (21), and because mutations of helices 3 and 5 also diminish binding to corepressors, it has been postulated that conformational changes in helix 12 could also play a role in blocking corepressor binding when the receptor binds ligand (23, 24). In further support of the notion that helix 12 can affect corepressor were the observations with several nuclear receptors that deletion (25) or mutation (10) of helix 12 increases corepressor binding. However, the boundaries of the corepressor interaction surface have not been described, and the precise structural role of helix 12 in destabilizing corepressor binding has not been elucidated. Helix 12 has mostly been considered to be passive in unliganded nuclear receptors and to serve its main role in formation of the coactivator-binding surface. However, it is also possible that this helix is important in the unliganded state.
Two receptor interaction domains (RIDs) have been described within both SMRT and N-CoR (5, 16, 22, 26), and a third has been described for N-CoR (27). All of these contain motifs of IxxII that are important for binding (23, 24, 25, 27). The separate RIDs exhibit specificity in their binding to different nuclear receptors (18, 28), but the basis for this specificity is unknown. The IxxII motifs are likely present within larger protein structural domains that bind to the receptors. Differences between these domains outside the consensus sequences could dictate specificity for interacting with nuclear receptors, although there is no direct evidence for this idea. It is also unclear to what extent the separate RIDs recognize the same or different surfaces on TR or other nuclear receptors.
In this study we applied scanning surface mutagenesis to identify TR
surfaces that interact with three fragments of N-CoR, each of which
contains one of the RIDs. We introduced mutations over the entire
surface of the TR LBD, using the x-ray structure of the hormone-bound
TR LBD (21) as a guide for placing mutations. Even though
corepressors bind unliganded TRs, we cautiously anticipated that use of
the liganded TR LBD structure would be suitable, because the available
evidence suggests that most of the overall fold of unliganded and
liganded structures would be similar. In comparative studies we used
x-ray structures of the unliganded human RXR (hRXR
) LBD
(29) and the human ER (hER) LBD bound to the
mixed-antagonist 4-hydroxytamoxifen (30), to place
mutations in these receptors. We used these mutated receptors in
glutathione-S-transferase (GST) pull-down assays to identify
the surface regions important for binding to N-CoR, and we used a
selection of the vectors encoding mutated TRs for mammalian cell
transfection to study the effect of the mutations on the TR unliganded
functions.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The TR area in which mutations had the strongest effects on N-CoR
binding was on the surface formed by helices 3 and 5, overlapping the
surface previously shown to bind coactivators (31). We
refer to this as Site 1. Mutations T277R, I280K, V284R, K288A, I302R,
and C309K all showed decreased in vitro binding by N-CoR
(Fig. 1A
). By contrast, mutations of
residues of helix 12 that form part of the coactivator-binding surface
(L454R and E457K) did not decrease binding to N-CoR, indicating that,
unlike the case with coactivator, the N-CoR-binding surface does not
include the outside of helix 12. Further, a natural mutation of the TR
that deletes helix 12 [F451X (32)] increased TR binding
to N-CoR by about 2.5-fold for both the RIDs I and II (Fig. 1A). Thus,
it appears either that N-CoR does not contact the residues of helix 12
or that such contact does not appreciably stabilize the
interaction.
|
, I280K/F451X and C309K/F451X). The finding
that residues obscured by helix 12 in the liganded TR contribute to the
binding surface provides support for the role hypothesized for helix 12
in destabilizing corepressor binding by demonstrating a mechanism:
helix 12 and corepressor compete for interactions with the same
residues.
The inferred corepressor-binding surface is partially obscured in the
x-ray structure model of the liganded TR LBD (Fig. 1C, left
panel), and therefore it is more easily viewed when the 11
C-terminal residues are not displayed, so as to show a structure that
might occur in the natural deletion mutation, F451X (Fig. 1C, right panel). This binding surface is relatively small and
is comprised of a cluster of hydrophobic residues I280, V283, V284,
I302, and C309) bordered by polar (T277 and T281) or charged (K288 and
K306) residues. The portion of this surface that overlaps with the
coactivator-binding surface includes residues V284, K288, I302, and
K306 (31).
Surface Structure of Site 1 Regulates RID Selectivity
Helices 3 and 5 are part of a region of
homology within the nuclear receptor family named "the signature
motif" (33). The fact that residues from these helices
appear to contact corepressor suggests that a large number of receptor
family members may bind corepressors. Binding of wild-type (WT) TR
occurs to all three N-CoR RIDs, but selectivity of other nuclear
receptors for separate corepressor RIDs has been reported (18, 28). The basis for such selectivity is unknown. We investigated
the effect of surface shape changes on selectivity by testing effects
of conservative mutations in which residues of helices 3 and 5 are
substituted with amino acids present in other nuclear receptors.
Although binding of the WT TR occurred to each of the three RIDs,
differential effects were found within this set of conservative
mutants, either specifically increasing (T281I, T281L, T281V, V284I,
and I302V) or decreasing (T281L, T281Q, T281V, V284I, and I302V)
binding to one of the individual RIDs (Fig. 1B). Other mutants
affected binding to all N-CoR RIDs similarly, resulting in molecules
with decreased (I280M, V283M, V284A, I302A, I302M, C309A, and C309W) or
little changed (V283I and L305I) binding (Fig. 1B). Because the tested
mutations are local and conservative, these results suggest that
variations in the shape of the interaction surface of different nuclear
receptors regulate their ability to bind corepressors and to define
different modes of binding.
Ligand Regulation of N-CoR Binding When Helix 12 Is Deleted
As demonstrated previously (25), we found that
deletion of the C-terminal helix of human RXR (equivalent to helix
12 of the TR) results in a large increase in N-CoR binding (Fig. 2, A and C). Strong binding occurred to
N-CoR RID I, but very little binding occurred to RID III and almost no
binding occurred to RID II (Fig. 2A), results that are consistent with
previous reports (28). Mutations of RXR residues
equivalent to those forming the coactivator-binding surface of TR helix
3 have been shown to decrease corepressor binding (25),
and we extended this finding to show that mutation of a residue of the
RXR (W305), whose equivalent in the hormone-bound TR (C309) would be
buried by helix 12, markedly decreases N-CoR binding (Fig. 2C). Thus it
appears that a surface of RXR similar to that of the TR is recognized
by N-CoR and that N-CoR binding to this surface is regulated by the
presence of helix 12 even in the absence of ligand.
|
could be obtained using
N-CoR RID I but not RID II (Fig. 2, B and D), but good binding occurred
with RID III (Fig. 2B). Furthermore, binding of RID I to the ER was
enhanced by deletion of helix 12 (Fig. 2D). However, binding of RID I
to the helix 12-deleted ER also required addition of the mixed
antagonist, tamoxifen (Fig. 2B). This ligand stimulation was not
observed for binding to RID III (Fig. 2B). Three mutations of the ER
(I358R, K362A, and V376R) showed diminished binding by the helix
12-deleted ER (Fig. 2B), thus confirming that N-CoR binding likely
occurs to a surface similar to Site I of the TR. These three mutations
are located in the portion of Site 1 where corepressor and coactivator
binding overlap. A fourth mutation of ER (W382K), which would occur in
the part of Site 1 that is buried by the hormone-bound helix 12, also
abolished N-CoR binding (data not shown), but ER W382 likely is also
important for the binding of tamoxifen (30). The ligand
requirement for N-CoR RID I binding by the ER in the absence of helix
12 implies that mechanisms in addition to a conformation change in
helix 12 exist for regulation of corepressor binding. Presumably,
ligand binding causes conformation changes in parts of the nuclear
receptor in addition to helix 12.
We found that the TR can also be regulated by ligand-induced
conformational mechanisms. Although addition of Triac blocks binding of
N-CoR to the WT TR LBD that contains helix 12, it stimulates binding of
N-CoR to the helix 12-deleted TR LBD (Fig. 2E). This effect required a
high concentration of Triac, consistent with the expected weak binding
affinity of Triac for the helix 12-deleted TR. We determined that the
F451X TR LBD binds with a dissociation constant
(Kd) of 100 nM in saturation
radioreceptor binding assays with
[125I]T3, and that this
binding is competed more effectively by T3
compared with reverse T3, and not competed at all
by tamoxifen (data not shown). Therefore, ligand-induced stimulation of
corepressor binding is likely a general property of nuclear receptors,
suggesting that pharmacological approaches may be more generally useful
than previously appreciated for regulating corepressor functions via
the stimulation of corepressor recruitment by nuclear receptors.
Mutations in Two Additional Surfaces of the TR LBD Influence N-CoR
Binding
To investigate the previously defined CoR box (4)
region, we mutated every surface residue on helix 1, and assayed these
for N-CoR binding. Confirming the previous report (4),
simultaneous mutation of three buried residues of helix 1 (A228, H229,
and T232, AHT) was defective in binding N-CoR RIDs I and II (Fig. 3A). Most of the surface mutations of
helix 1 had no effect on N-CoR binding, although one surface mutation
located at the beginning of helix 1 (W219K, Fig. 3C) had impaired
binding to RIDs I and II (Fig. 3A), and RID III (data not shown). This
suggests there is little contribution to N-CoR binding by most of the
helix 1 surface but indicates the region near W219 may regulate
binding.
|
), although not as strongly
as did some mutations of Site 1. Together with W219, these residues
define a concave surface running alongside helix 1 (Fig. 3C), which we
refer to as "Site 2." Site 2 contains hydrophobic, polar, and
charged residues that derive from distant locations in the TR linear
sequence. Because mutation of the surface-exposed residues of helix 1
that lie over the buried AHT mutation do not affect N-CoR binding but
do define a border of Site 2, effects of the AHT mutation could be due
indirectly to effects on Site 2.
Amino acids on helix 11 have also been implicated in corepressor
binding, because deletions of the terminal portion of helix 11 affect
both N-CoR and SMRT binding (16, 22). Helices 10 and 11
participate in formation of the surface for TR-TR homodimerization and
TR-RXR heterodimerization, and mutation L422R has the strongest dimer
destabilizing effect (34). Most surface mutations of
helices 10 and 11, including L422R, and a buried mutation of helix 11
(L428R) did not diminish binding to N-CoR RIDs I and II (Fig. 4A). Two mutations on the surface of
helix 11 (K424R and D427A) caused increases in binding to N-CoR RID II
(Fig. 4A). Two mutations of helix 10 (Q396R and L401R) caused weak
(average 25%) decreases, and one mutation of helix 11 (R429A) caused
stronger (60%) decreases in N-CoR binding (Fig. 4A). The Q396R, L401R,
and R429A mutations were also demonstrated to decrease the binding to
N-CoR RID III (data not shown). Whereas this surface influences N-CoR
binding, the structural pattern is discontinuous rather than clustered
(Fig. 4B). Residue R429, the most sensitive to mutation, has its side
chain pointed above helix 11 toward residue Q396 of helix 10. We refer
to these two residues as "Site 3." Residue L401 does not contact
residues 396 or 429; its side chain points above helix 10 toward where
the DBD could be oriented in the full-length TR. In Fig. 4B, residues
that give rise to decreased binding to all N-CoR RIDs are colored by
residue type. Residues K424 and D427, which increase binding to N-CoR
RID II, are colored gray; these two residues are directed
below helix 11, toward helix 8 (Fig. 4B).
|
). When the
amount of unliganded repression is plotted against the amount of N-CoR
binding for each TR mutant, a significant correlation is obtained, with
r2 values of 0.52, 0.59, and 0.62 for RIDs I, II,
and III, respectively (Fig. 5C). The mutations tested include those
from Sites I, II, and III, and therefore likely indicate that the
sensitivity to mutations for in vitro N-CoR binding have
functional significance in vivo. One mutation (L422R) had
strong effects in vivo without a correlated defect in the
in vitro binding assay. Because this mutation affects
receptor dimerization (34), this could suggest this
unliganded function requires in vivo formation of a TR dimer
or some other association at this site.
|
promoter linked to LUC. The unliganded
TR increased LUC expression from this promoter by 2.5-fold, and
mutations that impair N-CoR binding also were deficient in this
unliganded activation (Fig. 5B). There was a significant overall
correlation between defects in binding and this function
(r2 values of 0.44, 0.66, and 0.56 for RIDs I,
II, and III, respectively), although again, the L422R mutation was an
outlier (Fig. 5D). There was also a significant correlation when the
relative TRE repression and relative TSH promoter activation results
for the various mutants were compared (r2 value
of 0.81, plot not shown), consistent with both functions requiring
interactions with corepressor. Thus defects in N-CoR binding have
functional importance in vivo, and regulation of the TSH
promoter by unliganded TR may require receptor dimers or some other
partner using this surface.
Specificity of Conformation Defects by Mutation of Sites 2 and
3
It has been suggested that the AHT triple mutation of buried
residues of helix 1 may decrease corepressor binding indirectly through
a change in conformation of the unliganded TR (36). The
AHT mutations are near the mutation-sensitive Site 2 alongside helix 1,
and therefore mechanism(s) for decreasing N-CoR binding may be similar
for all these mutations. Residues of Sites 2 and 3 are on the surface
of the hormone-bound TR LBD (21), making it less likely
that mutations at these sites cause strong global structural changes.
However, no unliganded TR structure has been determined, and the
disposition of the residues of Sites 2 and 3 could be different in the
unliganded conformation required for corepressor binding. Therefore it
is not obvious whether Sites 2 and 3 represent alternate direct binding
sites for N-CoR or whether the effects of mutations at these sites
involve structural changes that indirectly affect N-CoR binding to Site
1.
To test whether mutational defects in N-CoR binding might be mediated
by global structural changes, we examined abilities of the AHT triple
mutation and TR mutations from Site 1 (I280M and I302R), Site 2
(W219K), and Site 3 (R429A) to function in other capacities (Fig. 6). Parts of Sites 1, 2, and 3 occur in
the TR LBD subdomain that does not bind hormone; therefore, as a
comparative control, mutation W418K of a partially buried residue of
helix 11 was created to intentionally destabilize this same subdomain.
The W418K mutation (displayed separately in Fig. 6) did not diminish
binding to N-CoR (Fig. 4A). All the mutants showed good binding of
[125I]T3 (Fig. 6A),
although all exhibited a decreased affinity relative to WT TR.
Therefore, each mutation appears to have some global effect. The worst
[125I]T3 binder was the
AHT mutation with a Kd of 600 pM.
|
), using excess hormone (50
µM). The I302R mutation, which participates in the p160
coactivator-binding surface (31), showed only background
binding. By contrast, the I280M and I280K (data not shown) mutations
that lie underneath helix 12 in the liganded TR conformation, and which
strongly block N-CoR binding to Site 1 (Fig. 1B), showed almost no
defect in GRIP1 binding. These results are consistent with a model in
which hormone binding causes I280 to be covered by helix 12, so that
the corepressor binding Site 1 is blocked, and the coactivator binding
surface is created. Thus, mutation of the buried part of the
corepressor surface allows the best discrimination between corepressor
and the coactivator binding by the TR. Mutations from Sites 2 and 3 and
the control mutation all showed similar partial defects in GRIP1
binding, implying that these mutations have similar global defects in
recruitment of coactivator. A similar pattern was obtained for hormone
activation by these mutant receptors in transfection assays (Fig. 6C);
the I302R mutant was inactive, but other mutants from Sites 2 and 3 and
the control W418K mutant, although capable of good activation, were
similarly slightly deficient. Thus, the mutations in Sites 2 and 3
exhibit indications of global structural changes, but no more so than
the control W418K mutation that had no effect on N-CoR binding.
Mutation of Site 2 Alters N-CoR Recognition by the TR LBD
Mutation of Site 1 diminished N-CoR binding to levels close to the
control GST protein, whereas the mutations of Sites 2 and 3 only
partially blocked binding. We asked whether binding that remains
in the W219K (Site 2) mutant could be blocked by secondary mutations of
Site 1 and found the secondary Site 1 mutations to be surprisingly
ineffective (N-CoR RID I; Fig. 7).
Although low binding was observed for some of the single Site 1
mutants, these same mutants showed higher binding when present as
double mutants coupled with the W219K (Site 2) mutation. Thus,
mutation of residue W219 may cause a conformation change that affects
Site 1 recognition. A search was made for LBD residues outside of Site
1 that would more effectively block N-CoR binding in combination as
double mutants with W219K. The most effective of these secondary
mutations was R429A, which nearly completely blocked the remaining
binding (Fig. 7 and data not shown). When the Site 1 mutations were
combined with the R429A (Site 3) mutation, there was either no change
(V284R), a smaller increase (I280M), or increased (I302R) binding,
indicating that recognition of Site 1 by R429A is somewhat altered, but
mostly intact.
|
). Deletion of
helix 12 caused a 5-fold increase in binding to N-CoR RID I, larger
than the 2.5-fold increase observed with full-length receptor, implying
that the N terminus and/or DBD affect binding. These effects may be
conformational because we could find no evidence of direct binding to
the N terminus or DBD using a mutation (A234X) that deletes the LBD
(data not shown). In the LBD the R429A (Site 3) mutation showed little
of the defect observed in the full-length TR, again implying that the N
terminus and/or DBD influence corepressor binding. In the LBD the W219K
(Site 2) mutation remained defective as for the full-length, and
deletion of helix 12 allowed little improvement, consistent with an
effect of Site 2 mutations to alter the recognition of Site 1. The
I302R (Site 1) mutation diminished binding of the LBD to background
levels, and deletion of helix 12 showed no improvement, consistent with
a requirement of residue I302 in contacting N-CoR RID I.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
LBD (21) as a structural model. The results imply a
complex pattern of interaction between N-CoR and the TR, as reflected
by the fact that we found not one, but three sites of mutational
sensitivity (Sites 1, 2, and 3, Fig. 9).
Site 1 overlaps partially the surface defined for coactivator binding
(31) but also extends underneath where helix 12 rests in
the ligand-bound conformation. Site 2 is separated from Site 1 by
helices 1 and 3 and runs along the side of helix 1; Site 3 is separated
from Site 1 by the turn between helices 9 and 10 and runs between
helices 10 and 11. Mutations in Site 1 nearly abolished binding, from
which we infer that Site 1 contacts corepressor directly. Mutations in
Sites 2 and 3 had partial effects, which could indicate that they form
weaker contacts to corepressor. Alternatively, Sites 2 and 3 could
function to define and maintain the presentation and stability of the
unliganded TR conformation required for corepressor binding.
|
The data demonstrate that helix 12 can regulate corepressor binding by nuclear receptors in the unliganded state through its affinity for Site 1. This is based on the observation that deletion of helix 12 in the unliganded TR increases corepressor binding by N-CoR RIDs I and II and, in the RXR, has a profound effect on corepressor binding (25), confirmed here. In some receptors, competition by helix 12 might be strong enough to preclude the use of corepressors as coregulators in the cell. However, it is conceivable these nuclear receptors interact with corepressors in vivo if mechanisms exist that reduce the competitive effect of helix 12, such as the ability of helix 12 to be drawn away from one receptor by other receptor molecules in a complex (39), or in vivo proteolysis of helix 12.
Although binding of corepressor to Site 1 requires displacement of
helix 12 from its liganded conformation, the mapping of Site 1
indicates that the extent of required displacement is not large.
Complete extension of helix 12, as reported for unliganded RXR
(29), is probably not required. Sliding of helix 12 down
along helix 3, as reported for unliganded PPAR (40), would
provide sufficient exposure, as illustrated in Fig. 9. Helix 12 could
also be uncoiled in the absence of ligand, because, even when
stabilized by ligand, the helix 12 of crystallized TR shows a B-factor
higher than the average for the entire LBD (21). It has
been predicted (24) that the IxxII motifs within
corepressors that interact with receptors (23, 25) extend
longer as amphipathic helices, compared with the LxxLL motifs of
coactivators. This prediction is consistent with our observation that
the corepressor-binding surface extends beyond that for
coactivator.
The residues of Site 1 are homologous between members of the nuclear receptor superfamily, deriving from the signature motif sequences between helices 3 and 5 (33). This observation suggests that corepressors might dock to Site 1 of various nuclear receptors more ubiquitously than previously thought. This is supported by our observation that mutation of the residues of Site 1 in the ER and RXR, including those under helix 12, also impaired corepressor binding. Interestingly, previous mutation of this region in the orphan nuclear receptor RevErb was found to impair corepressor binding (17). This receptor lacks helix 12, and the role of helix 12 in restricting corepressor binding was not derived from this study. Site 1 may be used generally, so long as it has a sufficient exposure to, and affinity for, the corepressor.
Despite similarities between different nuclear receptors in Site 1, receptors also show selectivity for the separate corepressor RIDs. Our results indicate that the shape of the Site 1 surface is one determinant of this selectivity. Whereas the helix 12 deleted unliganded RXR and tamoxifen-liganded ER bound RID I, they did not bind RID II, demonstrating that features other than helix 12 define selectivity. By contrast, TR exhibited equal binding to N-CoR RIDs I and II. We further observed that conservative changes in the residues of TR Site 1, such as occur naturally between different nuclear receptors, affected binding uniquely to N-CoR RIDs I, II, and III, either decreasing or increasing the affinity of receptor binding. Thus, the residues of Site 1 partially determine the selectivity. However, none of the single point mutations of Site 1 conferred selectivity as large as that observed for the RXR or ER, and no mutation reversed the preference of TR to bind better to RID III than I or II, indicating that either more than one mutation of Site 1 or contacts to or changes outside of Site 1 are required to dictate the complete selectivity observed when the natural receptors are compared. When previous reports on receptor selectivity are compared with the current results, it should be noted that differences often exist in the boundaries defined for the LBD-interacting RIDs of N-CoR (27), and that additional contacting surfaces of receptors and corepressors may yet be elucidated in the future (41) that could influence binding results in an assay-specific fashion.
Our study adds insights about the CoR box, which was originally reported to be the TR site for interactions with corepressors (4). The current results extend those of previous reports (24, 42), which found that mutations of the surface residues of helix 1 lying directly above the CoR box did not affect corepressor binding, indicating that the CoR box, as postulated, does not bind corepressor. We extended these studies by testing mutations at every surface residue of helix 1 and found one mutation (W219K) that diminished corepressor binding. Residue W219 on helix 1 forms part of surface Site 2. Thus, the previous mutations supporting the existence of the CoR box may have perturbed the receptor structure in a way that affects Site 2.
The mechanisms for decreased corepressor binding due to mutations of Sites 2 and 3 remain unclear. Some residues of Sites 2 and 3 are unique to TR, but it is notable that residues D366 and R429 are relatively conserved within some nuclear receptor family members (33), consistent with roles in either supporting the TR structure or providing protein-binding interfaces. The Site 2 and 3 mutants showed only small defects in both ligand binding and coactivator binding that are comparable to defects of another TR mutant (W418K) that is not defective in N-CoR binding. These results are similar to previous reports that natural TR mutants found in patients with the syndrome of Resistance to Thyroid Hormone (43, 44, 45, 46, 47) generally show some ligand-binding defects, but no loss of corepressor binding. Thus we conclude that if the effects of mutations at Sites 2 or 3 on N-CoR binding are due to structural perturbations, these perturbations are specific. More evidence supporting a mechanism whereby Site 2 and 3 mutations are required to specifically maintain the unliganded conformation essential for presentation of Site 1 as the primary N-CoR interaction surface was revealed by the observations that mutations of Sites 2 or 3 can alter the mutational sensitivity of Site 1, and that under some conditions ligand can reverse the effects of the Site 2 mutation.
Although conformation mechanisms likely contribute to the
corepressor-binding defects of Sites 2 and 3, a role for these surfaces
as direct contact sites should also be considered. Site 2 is separated
from Site 1 by one-third the circumference of the LBD (
30 Å),
whereas Site 3 is closer (10 Å distance between residues R429 and
C309). The TR LBD has a molecular mass of 32 kDa, whereas the fragments
of N-CoR used to make the fusion proteins for GST pull-down experiments
are 9.1, 8.8, and 16.4 kDa for RIDs I, II, and III,
respectively. It seems unlikely the smaller N-CoR fragments
could contact Sites 1 and 2 simultaneously without making contacts to
the intervening residues of helices 1 and 3 simultaneously. By
contrast, it seems possible for the N-CoR fragments to contact Sites 1
and 3 simultaneously. Such a model would be consistent with our finding
that the N-CoR binding activity remaining in the partially defective
W219K (Site 2) mutant requires the integrity of residue R429A (Site 3)
more than it requires the residues of Site 1. The N-CoR binding
activity remaining in the R429A (Site 3) mutant is abolished by either
the W219K (Site 2) mutation or by some of the Site 1 mutations. Thus,
the Site 3 mutations could act through both conformational and direct
binding mechanisms.
Conceivably, Sites 2 and 3 could be interaction sites for other molecules that modulate N-CoR binding through allosteric mechanisms. The [35S]-labeled TRs used for the binding were produced using reticulocyte lysates, and therefore numerous candidates for putative allosteric regulators would have been present in our pull-down assays. The effects of the Site 2 and 3 mutations primarily appear not to be mediated through effects on the N terminus or DBD, because deletion of these domains generally did not remove the effects of the Site 2 or 3 mutations. However, the R429A (Site 3) mutation failed to affect the binding to RID I in the context of the LBD, so some role of the N terminus or DBD cannot be ruled out.
Effects of the mutations on the in vivo functions of the full-length unliganded TR were assayed using both T3-activated and T3-repressed promoters. All mutations of Sites 1, 2, and 3 that diminished in vitro N-CoR binding also caused correlated defects in unliganded repression and unliganded activation, and these two functions were also highly correlated when compared across the set of mutants. These results imply that the binding defects measured in vitro are also responsible for defects in corepressor binding and action measured in vivo. Interestingly, one mutation that did not affect N-CoR binding in vitro, L422R, strongly diminished the unliganded TR in vivo functions. L422R strongly affects stability of both TR dimers and heterodimers with RXR (34). N-CoR binding by DNA-bound TRs has been reported to require homodimers (48). Thus it is possible that the in vivo defects observed with L422R indicate that formation of a TR dimer is essential for corepressor to function at these promoters. As the TSH promoter contains no strong binding site for the TR, the actions of such a putative dimer would likely occur off the promoter DNA in this case.
N-CoR binding by the helix 12-deleted TR and ER was surprisingly stimulated by ligand binding. N-CoR binding by the ER with helix 12 deleted was barely detectable unless tamoxifen was bound. In the case of TR with helix 12 deleted, the agonist Triac stimulated N-CoR binding by 30%. These observations provide evidence that influences on the receptor extrinsic to helix 12 can regulate N-CoR binding. Tamoxifen releases ER from its associated chaperone proteins that are present in the reticulocyte lysates used to produce the radiolabeled receptors (49). Thus, this release could free the receptor from inhibitory influences of these proteins and thereby stimulate N-CoR binding. However, the unliganded TR is not known to associate with proteins other than corepressors. Thus, an alternative explanation is that tamoxifen and Triac could stabilize the LBD in a manner that favors N-CoR binding. The result that binding by the W219K/F451X mutant to N-CoR RID I was stimulated 6-fold by Triac suggests the defect caused by the W219K (Site 2) mutation can be appreciably corrected by addition of ligand. Thus there could be a continuum of dynamic states of the TR with different capacities of corepressor binding. In such a model, specifically located mutations, such as those of Site 2, could perturb the receptor conformation to disable binding, but other influences, such as ligand binding or, conceivably, allosteric effects with other domains of the TR or with TR dimerization partners, could stabilize conformations to improve binding.
The current studies also suggest pharmacological approaches to modulate selectively nuclear receptor functions through regulating corepressor binding. Ligands that bind to the combined coactivator and corepressor surface might impair or stimulate both corepressor and coactivator action. Other ligands that bind to the corepressor-binding surfaces might selectively impair or stimulate corepressor function. The fact that mutations such as C309K and I280M have a much more profound affect on corepressor than coactivator binding, despite their localization in a region where they could impair packing of helix 12, implies that such an approach might be successful. Ligands that bind in the hormone-binding cavity might also be used for selective regulation. In addition to the tamoxifen-stimulated binding of corepressors, it has been reported that different retinoids vary in their capacity to release corepressors (50). These results, coupled with our finding that ligands can enhance corepressor binding by helix 12-deleted TRs, suggest that in some contexts ligands that range in effect from corepressor recruitment to release for the same receptor may be possible.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
, and the pSG5 vector was used for expression of the
full-length human ER (hER). Mutations within these nuclear
receptor-encoding sequences were created using the methods described in
the quick-change kit (Stratagene, La Jolla, CA). The same
mutagenesis methods were used in protocols to engineer the
Escherichia coli expression vectors encoding fragments of
N-CoR and TR. For the GST-N-CoR RID I, the sequences encoding amino
acids 2,231–2,321 were cloned in frame to the C terminus of the 26-kDa
GST protein of the GEX-2T vector (Pharmacia Biotech,
Piscataway, NJ) between modified linker sequences encoding His-Met and
encoding Val-Ala-His-His-His-His-His-His. An NdeI site
encodes the His-Met and a SalI site encodes the Val-Ala. The
appropriate NdeI and SalI sites were introduced
into the mouse N-CoR cDNA sequence using sequential mutagenesis steps.
The same approach was used to make the GST-N-CoR RID II (encoding
residues 2,034–2,114) and GST-N-CoR RID III (encoding residues
1,888–2,031). Verification of the target sequences as well as the
flanking sequences was performed using both Sequenase kits
(Stratagene), and by automated DNA sequence (UCSF Cancer
Center Sequencing Facility).
GST Pull-Down Assay
The vectors encoding the GST-N-CoR-His fusion proteins of the
three N-CoR RIDs were used to transform the bacterial strain BL21,
using selection with ampicillin. One-liter cultures were grown using
2x LB medium to an OD600 of 0.6, and expression
was induced with room temperature shaking by addition of 0.5
mM isopropyl-ß-D-thiogalactopyranoside, for
3 h. Cells were harvested by centrifugation and after storage at
-80 C, extracted with 20 ml TST buffer [50 mM Tris, pH
7.5, 150 mM NaCl, 0.05% Tween-20, with 0.02%
monothioglycerol and 20 µM phenylmethylsulfonyl fluoride
(PMSF) added]. The lysate was centrifuged for 1 h at 17,000 rpm
in a JA20 rotor (Beckman Coulter, Inc., Palo Alto, CA),
and the lysates were mixed for 1 h at 4 C with 0.4 ml Talon Metal
affinity resin (CLONTECH Laboratories, Inc., Palo Alto,
CA) in the presence of 2 mM imidazole. Beads were
collected by low-speed centrifugation and washed twice batchwise with
10 ml buffer S (50 mM NaPO4, pH 8,
300 mM NaCl, and 10% glycerol, with 2 mM
imidazole added). Beads were then transferred to a 5 ml Econo column
and washed with 2 ml buffer S containing 2 mM imidazole.
The GST-N-CoR-His proteins were then eluted with 2 ml buffer S
containing 100 mM imidazole. Metal affinity-purified
GST-N-CoR His-tagged fragment (1.5 mg) was mixed for 1 h with 150
µl of glutathione Sepharose (Pharmacia Biotech), then
washed three times with 1 ml TST buffer with 1 mM
dithiothreitol added, and brought up to 150 µl of a 50% slurry,
typically containing 5 mg/ml protein.
For studies of binding to the GST-N-CoRs, [35S]-Metlabeled nuclear receptors were made using 1 µg of each receptor-encoding plasmid in 20 µl reactions using TNT kits (Promega Corp., Madison, WI). For each experiment, equivalent amounts (15 fmol) of labeled receptors were diluted in 150 µl binding buffer (20 mM HEPES, pH 7.9, 150 mM KCl, 25 mM MgCl2, 10% glycerol, 0.1% Triton X-100, 0.1% NP-40, and 0.01% dithiothreitol), of which duplicate 10-µl portions (20% of the binding reaction inputs) were quantified by SDS-PAGE and phosphorimaging (Molecular Dynamics, Inc., Sunnyvale, CA). Binding buffer (100 µl) containing 4 µl of the glutathione Sepharose-bound GST-N-CoR-His was added to duplicate 50-µl portions of the diluted labeled receptors and mixed by mechanical inverting for 2 h at 4 C. The beads were washed three times with 1 ml binding buffer with brief microcentrifugation in between. The washed beads were boiled 3 min in SDS sample buffer and quantified by SDS-PAGE and phosphorimaging.
Mammalian Cell Culture and Transfection
The unliganded functions of the TR were analyzed by comparing
the effects of a CMX vector encoding the WT and mutant receptors with
the effects observed for the empty parent CMX vector. For unliganded
repression studies, a LUC reporter plasmid was cotransfected that
contained two copies of the F2 TRE from the chicken lysozyme gene
(35) linked to the herpes simplex virus TK promoter
sequences extending from -109 to +48 of the transcription start site.
For these studies of the F2-TK109 promoter, CV-1 cells were harvested
from DMEM-H21 supplemented with 10% FBS and plated at 7,000
cells per well. The cells were washed 2 h later with 2 ml PBS, and
new medium was added containing serum stripped of thyroid hormone using
Dowex AG X-8 ion exchange resin (Bio-Rad Laboratories, Inc., Hercules, CA) (51). One and one-half hours
later, 0.5 ml suspension of DNA vectors precipitated with calcium
phosphate was added [1.7 µg LUC reporter, 80 ng CMX, or CMX-TR
expression vector, and 500 ng control ß-galactosidase vector]. After
15 h cells were washed twice with 2 ml DME-H21 and fresh DME-H21
containing stripped serum was added. After an additional 24 h,
cells were harvested in 200 ml lysis buffer and assayed for LUC (kit
from Promega Corp.) and ß-galactosidase (kit from
Tropix, Inc., Bedford, MA) activities.
For unliganded activation studies, a reporter consisting of the TSH
gene promoter linked to LUC (3) was transfected by the
calcium phosphate method (3) into TSA-201 cells and grown
in DMEM (Nikken Biomedical Laboratory, Kyoto, Japan) with 10%
Dowex resin-stripped FBS, penicillin (100 U/ml), and streptomycin (100
µg/ml). After exposure to the calcium phosphate-DNA precipitate for
8 h, DMEM with 10% Dowex resin-stripped FBS was added. Cells were
harvested after 40 h for measurements of LUC activity.
For T3 activation studies, a LUC reporter plasmid was used containing two copies of a F2 thyroid response element linked to the herpes simplex virus TK -109/+48 promoter. HeLa cells were cultured and transfected as described previously (31).
T3 Binding Assay
Affinity measurements for
125I-T3 binding to hTRß
mutant proteins produced by in vitro translation were
performed as described previously (52).
|
FOOTNOTES |
|---|
1 Current address for A.M., H.N., and B.L.W.: Plexxikon, Inc., 91
Bolivar Street, Berkeley, California 94710. 
2 J.D.B. has proprietary interests in and serves as a consultant and
Deputy Director to Karo Bio AB, which has commercial interests in this
area of research.
Abbreviations: DBD, DNA-binding domain; GRIP, GRinteracting protein; GST, glutathione-S-transferase; h, human; LBD, ligand-binding domain; LUC, luciferase; N-CoR, nuclear receptor corepressor; RID, receptor interaction domain; SMRT, silencing mediator of retinoid and thyroid receptor; TK, thymidine kinase; TRE, thyroid response element; WT, wild type.
Received for publication February 26, 2001. Accepted for publication October 3, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
gene by the
thyroid hormone receptor. J Biol Chem 274:22345–22353 interacting protein. J Biol
Chem 274:15901–15907
) mutant is a
constitutive repressor and inhibits PPAR-mediated adipogenesis.
J Biol Chem 275:5754–5759 and RVR: physical
association is dependent on the E region of the orphan receptors.
Nucleic Acids Res 24:4379–4386. Nature 375:377–382[CrossRef][Medline]
4 activation domain of the thyroid hormone receptor is
required for release of a putative corepressor(s) necessary for
transcriptional silencing. Mol Cell Biol 15:76–86[Abstract]
. Nature 395:137–143[CrossRef][Medline]
. J Biol Chem 274:2885–2892 1 thyroid hormone receptor ligand binding domain in
Escherichia coli and the use of a ligand-induced
conformation change as a method for its purification to
homogeneity. Protein Expr Purif 6:363–370[CrossRef][Medline]
This article has been cited by other articles:
 |
|
 |
|
  T. Tagami, H. Yamamoto, K. Moriyama, K. Sawai, T. Usui, A. Shimatsu, and M. Naruse The retinoid X receptor binding to the thyroid hormone receptor: relationship with cofactor binding and transcriptional activity J. Mol. Endocrinol., May 1, 2009; 42(5): 415 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Araki, H. Ying, X. G. Zhu, M. C. Willingham, and S. Y. Cheng Distinct Dysregulation of Lipid Metabolism by Unliganded Thyroid Hormone Receptor Isoforms Mol. Endocrinol., March 1, 2009; 23(3): 308 - 315. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tagami, H. Yamamoto, K. Moriyama, K. Sawai, T. Usui, A. Shimatsu, and M. Naruse A Selective Peroxisome Proliferator-Activated Receptor-{gamma} Modulator, Telmisartan, Binds to the Receptor in a Different Fashion from Thiazolidinediones Endocrinology, February 1, 2009; 150(2): 862 - 870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Sanchez-Martinez, A. Zambrano, A. I. Castillo, and A. Aranda Vitamin D-Dependent Recruitment of Corepressors to Vitamin D/Retinoid X Receptor Heterodimers Mol. Cell. Biol., June 1, 2008; 28(11): 3817 - 3829. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kressler, M. B. Hock, and A. Kralli Coactivators PGC-1beta and SRC-1 Interact Functionally to Promote the Agonist Activity of the Selective Estrogen Receptor Modulator Tamoxifen J. Biol. Chem., September 14, 2007; 282(37): 26897 - 26907. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Jonas, N. Varlakhanova, F. Hayakawa, M. Goodson, and M. L. Privalsky Response of SMRT (Silencing Mediator of Retinoic Acid and Thyroid Hormone Receptor) and N-CoR (Nuclear Receptor Corepressor) Corepressors to Mitogen-Activated Protein Kinase Kinase Kinase Cascades Is Determined by Alternative mRNA Splicing Mol. Endocrinol., August 1, 2007; 21(8): 1924 - 1939. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Heldring, T. Pawson, D. McDonnell, E. Treuter, J.-A. Gustafsson, and A. C. W. Pike Structural Insights into Corepressor Recognition by Antagonist-bound Estrogen Receptors J. Biol. Chem., April 6, 2007; 282(14): 10449 - 10455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Lupien, M. Jeyakumar, E. Hebert, K. Hilmi, D. Cotnoir-White, C. Loch, A. Auger, G. Dayan, G.-A. Pinard, J.-M. Wurtz, et al. Raloxifene and ICI182,780 Increase Estrogen Receptor-{alpha} Association with a Nuclear Compartment via Overlapping Sets of Hydrophobic Amino Acids in Activation Function 2 Helix 12 Mol. Endocrinol., April 1, 2007; 21(4): 797 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Flamant, K. Gauthier, and J. Samarut Thyroid Hormones Signaling Is Getting More Complex: STORMs Are Coming Mol. Endocrinol., February 1, 2007; 21(2): 321 - 333. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Padron, L. Li, E. M. Kofoed, and F. Schaufele Ligand-Selective Interdomain Conformations of Estrogen Receptor-{alpha} Mol. Endocrinol., January 1, 2007; 21(1): 49 - 61. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Flamant, J. D. Baxter, D. Forrest, S. Refetoff, H. Samuels, T. S. Scanlan, B. Vennstrom, and J. Samarut International Union of Pharmacology. LIX. The Pharmacology and Classification of the Nuclear Receptor Superfamily: Thyroid Hormone Receptors Pharmacol. Rev., December 1, 2006; 58(4): 705 - 711. [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Santos, V. Afonso, G. B. Barra, M. Togashi, P. Webb, F. A. R. Neves, N. Lomri, and A. Lomri Negative Regulation of Superoxide Dismutase-1 Promoter by Thyroid Hormone Mol. Pharmacol., September 1, 2006; 70(3): 793 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Moran and G. Jimenez The tailless nuclear receptor acts as a dedicated repressor in the early Drosophila embryo. Mol. Cell. Biol., May 1, 2006; 26(9): 3446 - 3454. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Togashi, P. Nguyen, R. Fletterick, J. D. Baxter, and P. Webb Rearrangements in Thyroid Hormone Receptor Charge Clusters That Stabilize Bound 3,5',5-Triiodo-L-thyronine and Inhibit Homodimer Formation J. Biol. Chem., July 8, 2005; 280(27): 25665 - 25673. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Wang, C. Zhang, A. Marimuthu, H. I. Krupka, M. Tabrizizad, R. Shelloe, U. Mehra, K. Eng, H. Nguyen, C. Settachatgul, et al. The crystal structures of human steroidogenic factor-1 and liver receptor homologue-1 PNAS, May 24, 2005; 102(21): 7505 - 7510. [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]
|
|
|
|
|
|
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]
|
|
|
|
|
|
J. Jyrkkarinne, B. Windshugel, J. Makinen, M. Ylisirnio, M. Perakyla, A. Poso, W. Sippl, and P. Honkakoski Amino Acids Important for Ligand Specificity of the Human Constitutive Androstane Receptor J. Biol. Chem., February 18, 2005; 280(7): 5960 - 5971. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Farboud and M. L. Privalsky Retinoic Acid Receptor-{alpha} Is Stabilized in a Repressive State by Its C-Terminal, Isotype-Specific F Domain Mol. Endocrinol., December 1, 2004; 18(12): 2839 - 2853. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kawai, S. Sasaki, H. Morita, T. Ito, S. Suzuki, H. Misawa, and H. Nakamura Unliganded Thyroid Hormone Receptor-{beta}1 Represses Liver X Receptor {alpha}/Oxysterol-Dependent Transactivation Endocrinology, December 1, 2004; 145(12): 5515 - 5524. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Buchholz, A. Tomita, L. Fu, B. D. Paul, and Y.-B. Shi Transgenic Analysis Reveals that Thyroid Hormone Receptor Is Sufficient To Mediate the Thyroid Hormone Signal in Frog Metamorphosis Mol. Cell. Biol., October 15, 2004; 24(20): 9026 - 9037. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ishii, M. Yamada, T. Satoh, T. Monden, K. Hashimoto, N. Shibusawa, K. Onigata, A. Morikawa, and M. Mori Aberrant Dynamics of Histone Deacetylation at the Thyrotropin-Releasing Hormone Gene in Resistance to Thyroid Hormone Mol. Endocrinol., July 1, 2004; 18(7): 1708 - 1720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Wang, J. A. Blackford Jr., L.-N. Song, Y. Huang, S. Cho, and S. S. Simons Jr. Equilibrium Interactions of Corepressors and Coactivators with Agonist and Antagonist Complexes of Glucocorticoid Receptors Mol. Endocrinol., June 1, 2004; 18(6): 1376 - 1395. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Borngraeber, M.-J. Budny, G. Chiellini, S. T. Cunha-Lima, M. Togashi, P. Webb, J. D. Baxter, T. S. Scanlan, and R. J. Fletterick Ligand selectivity by seeking hydrophobicity in thyroid hormone receptor PNAS, December 23, 2003; 100(26): 15358 - 15363. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Benko, J. D. Love, M. Beladi, J. W. R. Schwabe, and L. Nagy Molecular Determinants of the Balance between Co-repressor and Co-activator Recruitment to the Retinoic Acid Receptor J. Biol. Chem., October 31, 2003; 278(44): 43797 - 43806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. U. Agoulnik, W. C. Krause, W. E. Bingman III, H. T. Rahman, M. Amrikachi, G. E. Ayala, and N. L. Weigel Repressors of Androgen and Progesterone Receptor Action J. Biol. Chem., August 15, 2003; 278(33): 31136 - 31148. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. G. Lehmann, J.-M. Wurtz, J.-P. Renaud, P. Sassone-Corsi, and E. Lalli Structure-function analysis reveals the molecular determinants of the impaired biological function of DAX-1 mutants in AHC patients Hum. Mol. Genet., May 1, 2003; 12(9): 1063 - 1072. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Fischer, S. M. G. Dias, M. A. M. Santos, A. C. Alves, N. Zanchin, A. F. Craievich, J. W. Apriletti, J. D. Baxter, P. Webb, F. A. R. Neves, et al. Low Resolution Structures of the Retinoid X Receptor DNA-binding and Ligand-binding Domains Revealed by Synchrotron X-ray Solution Scattering J. Biol. Chem., April 25, 2003; 278(18): 16030 - 16038. [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]
|
|
|
|
|
|
B. Farboud, H. Hauksdottir, Y. Wu, and M. L. Privalsky Isotype-Restricted Corepressor Recruitment: a Constitutively Closed Helix 12 Conformation in Retinoic Acid Receptors {beta} and {gamma} Interferes with Corepressor Recruitment and Prevents Transcriptional Repression Mol. Cell. Biol., April 15, 2003; 23(8): 2844 - 2858. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Webb, P. Nguyen, and P. J. Kushner Differential SERM Effects on Corepressor Binding Dictate ERalpha Activity in Vivo J. Biol. Chem., February 21, 2003; 278(9): 6912 - 6920. [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]
|
|
|
|
|
|
B. R. Huber, M. Desclozeaux, B. L. West, S. T. Cunha-Lima, H. T. Nguyen, J. D. Baxter, H. A. Ingraham, and R. J. Fletterick Thyroid Hormone Receptor-{beta} Mutations Conferring Hormone Resistance and Reduced Corepressor Release Exhibit Decreased Stability in the N-Terminal Ligand-Binding Domain Mol. Endocrinol., January 1, 2003; 17(1): 107 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.-J. Huang, J. D. Norris, and D. P. McDonnell Identification of a Negative Regulatory Surface within Estrogen Receptor {alpha} Provides Evidence in Support of a Role for Corepressors in Regulating Cellular Responses to Agonists and Antagonists Mol. Endocrinol., August 1, 2002; 16(8): 1778 - 1792. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Ghosh, X. Yang, A. Zhang, M. H. Lambert, H. Li, H. E. Xu, and J. D. Chen Interactions that determine the assembly of a retinoid X receptor/corepressor complex PNAS, April 30, 2002; 99(9): 5842 - 5847. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Ghosh, X. Yang, A. Zhang, M. H. Lambert, H. Li, H. E. Xu, and J. D. Chen Interactions that determine the assembly of a retinoid X receptor/corepressor complex PNAS, April 30, 2002; 99(9): 5842 - 5847. [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 |
H12). In all cases, duplicate
samples are performed to allow the average values to be quantified
after phosphorimaging. B, Binding of [35S]-labeled hER
