| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Lilly Research Laboratories, Department of Gene Regulation, Bone and Inflammation Research, Eli Lilly & Co., Indianapolis, Indiana 46285
Address all correspondence and requests for reprints to: Dr. Lan Ko, Department of Gene Regulation, Bone & Inflammation Research, Lilly Corporate Center, Building 98/C, Drop Code 0434, Indianapolis, Indiana 46285. E-mail: kol{at}lilly.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. Transient
transfection studies further confirmed that the LXXLL-binding affinity
correlates with TRBP transcriptional activity. Consistent with the
structural modeling, an E380G substitution within ER altered the
binding to TRBP mutants, demonstrating the direct contact between TRBP
S884 and ER E380, which is a residue that distinguishes receptor
subclasses. Furthermore, S884 can be phosphorylated by MAPK in
vitro, an event that significantly altered the binding of TRBP
to ER and suggests a potential mechanism for regulatory interaction. As
the differential recruitment of TRBP to ER and ERß may rely on
S884, our finding provides insight into estrogen signaling and may lead
to the development of therapeutic receptor-selective peptide
antagonists. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Crystal structures of the LBD of many nuclear receptors have been
resolved (36, 37, 38, 39, 40, 41). The coactivator LXXLL motif binding
site of ligand-bound ER is a shallow hydrophobic groove on the
surface of the LBD primarily formed by helices 3, 4, 5, and
12 (36). In antagonist-bound ER, the
coactivator-binding groove is obscured by unique positioning of helix
12. Alignment of helix 12 in a number of nuclear receptors with many
coactivator LXXLL peptides revealed that the hydrophobic residues in
helix 12 closely resemble the leucines in the LXXLL and appear to mimic
functionally the LXXLL motif of the coactivator (37).
Thus, helix 12 functions as an autoinhibitory peptide for
antagonist-bound nuclear receptors.
It has been shown that the distinct LXXLL motifs display varying degrees of nuclear receptor selectivity, which appear to be specified by the amino acid residues immediately flanking the LXXLL motifs (42, 43, 44, 45, 46, 47). The biological significance of selective interactions between various nuclear receptors and coactivators is still largely unclear; however, selective interaction in vivo may play a role in tissue- and cell-specific effects (48). In this regard, characterization of the mechanism(s) of coactivator selectivity is important for understanding coactivator function in nuclear receptor-mediated gene regulation.
The coactivator TRBP has been shown to interact with a number of
nuclear receptors in a ligand-dependent manner and enhance nuclear
receptor-mediated transcriptional activation. TRBP is a high mol wt
(2,063 amino acids) coactivator that associates with both CBP/p300 and
the DRIP complexes (30, 31, 33). TRBP also
coactivates multiple transcription factors including AP-1 and
NF
B (30) and its gene has been demonstrated to be
amplified in human breast cancers (31, 35). An
RRM-containing coactivator CoAA was recently shown to interact
with the carboxyl terminus of TRBP (49). In addition, the
carboxyl terminus of TRBP has also been shown to interact with nuclear
DNA-dependent protein kinase and poly (ADP-ribose) polymerase
complexes that are implicated in transcriptional regulation
(30). Interestingly, a single LXXLL motif is required for
the interaction of TRBP with TR, RXR, ER, and many other receptors.
However, the relative selectivity of these interactions between TRBP
and these receptors has not been characterized. To define the residues
flanking the TRBP LXXLL motif responsible for nuclear receptor
selectivity, we performed random mutagenesis of specific amino acids at
positions predicted to contact the surface of the nuclear receptor LBD.
A mutation at S884 at the -3 position of the TRBP LXXLL motif was
identified as a residue with the ability to alter TRBP’s selectivity
for ER vs. ERß both in vitro and in
vivo. Mutations at this residue also dramatically increased TRBP
binding to TR and RXR. We also show that the binding activities of TRBP
mutants to nuclear receptors correlates with their transcriptional
coactivation activities. Our results suggest that the interaction of
nuclear receptors with TRBP can be selective and may be regulated.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
To assess the importance of residues flanking the TRBP LXXLL motif in
determining receptor selectivity, we performed random mutagenesis of
TRBP (amino acids 714-1242) in which the -3, -4, +7, and +8 positions
near the LXXLL motif were mutated. These four residues were chosen
because the LXXLL motif is an -helical structure in which the -3,
-4, +7, and +8 positions are most likely aligned in the same
orientation as the conserved hydrophobic leucines. Hence, they are
likely to have direct contact with the receptor LBD. Degenerate PCR
primers introduced mutations randomly at these four specific amino acid
positions, and each mutated clone was screened for in vitro
nuclear receptor binding activity.
As shown in Fig. 1A
, 36 mutated TRBP
(714–1,242) clones were screened for binding of the encoded proteins
to the LBDs of TRß, RXR, ER, and ERß in the presence of their
respective ligands. Compared with the wild-type control, most of the
mutants displayed decreased binding, probably because of alteration of
the natural coactivator protein conformation. By calculation,
approximately 17% of the mutations would introduce stop codons and
fail to produce full-length proteins. This was detected by the lack of
an input band in the assay and was verified by sequencing of the
mutations (Figs. 1 and 2A). Our interest
was primarily focused on the mutations that display increased binding
activity. Among the positive clones, clone 6 had a surprisingly
significant increase in binding to TR and RXR and also had marked
increased binding to ERß but dramatically decreased binding to ER.
Interestingly, clone 33 had a similar pattern as clone 6, in that the
binding to TRß, RXR, and ERß was selectively higher than ER,
but to varying degrees. Sequence analysis of the clones revealed that a
tyrosine residue at position -3 contributed to selectivity of TRBP
binding of TRß, RXR, and ERß vs. ER (Fig. 2A). In
particular, S884 at position -3 in clones 6 and 33 was replaced by a
tyrosine. By contrast, most other mutants reduced TRBP-receptor
interaction. Of note, clone 6 had the least number of alterations at
other amino acid positions (S884Y and T883S), suggesting that the S884
position may significantly contribute to selective receptor
binding.
|
|
). Relative to the LXXLL
motifs of other coactivators, the LXXLL motif of TRBP is unique. It
contains a leucine at -1 position and a proline at the -2 position.
Consistent with previous reports, these characteristics lead to greater
binding to TRß (44, 45). As shown in Fig. 2B, the TRBP
LXXLL (882–895) motif structure was adapted and computationally
engineered based on the structure of the PPAR
AF-2 helix, which
aligns with the coactivator LXXLL motifs and has a leucine at the -1
and a proline at the -2 positions (37). The orientation
of the side chain of S884 in TRBP, due to Pro885, was found to align
with an orientation similar to that of the other four hydrophobic
leucines, L886, L887, L890, and L891 (Fig. 2A). The conformation of the
TRBP LXXLL motif suggested that S884 might be involved in receptor
binding, and this is consistent with our finding that mutation of S884
alters the binding of TRBP to the LBD.
S884Y Differentiates the Binding of TRBP with ER and ERß,
TRß, and RXR in Vitro
To verify our data obtained in the random
mutagenesis screen, the TRBP protein (714–1,242) with a S884Y mutation
was tested in in vitro binding assays. Binding of fusion
proteins consisting of glutathione-S-transferase (GST) fused
to the LBDs of TRß, RXR, ER, and ERß to wild-type TRBP and
the S884Y mutant were compared in the presence and absence of each
cognate ligand (Fig. 3). Consistent with
the observations from the screen, the S884Y mutant had a significant
increase in binding to TRß, RXR, and ERß, but decreased
binding to ER. Notably, the increase was strictly ligand dependent
for TRß and RXR, while S884Y increased ERß ligand-independent
binding. These results confirm that the S884 position plays an
essential role in determining receptor selectivity in
vitro.
|
and ERß using
various ER ligands including E2, diethylstilbestrol (DES), estriol,
genestein, LY 117018, 4- hydroxytamoxifen, and ICI 182,780. The
S884Y mutant of TRBP showed selective binding to ERß vs.
ER, although it displayed different profiles with each ligand (Fig. 4). In particular, the S884Y mutant
favored estriol-bound ERß over E2- and DES-bound ERß. This
selectivity for estriol was confirmed in transfection and peptide
binding assays. Genestein has been previously shown to exhibit
selectivity for ERß (50). The S884Y appeared to increase
genestein’s selectivity for ERß over ER, suggesting that the S884
residue may play a role in determining ER-coactivator complex
responsiveness to various ligands. All together, these data further
indicate that S884 plays an essential role in establishing TRBP-ER
isoform selectivity.
|
and
ERß in Vivo- or ERß LBD-VP16 fusions. The results suggest
that the S884Y mutation leads to increased binding of TRBP with ERß
and decreased binding with ER (Fig. 5). However, the ligand-independent
interaction with ERß was also significantly increased in the S884Y
mutant. The S884E mutant, which has a negatively charged side chain,
was not able to mimic the effect of the hydrophobic tyrosine and failed
to interact. Alanine mutations (LA) of the conserved leucines in the
LXXLL motif abolished TRBP interaction with the ER-LBD and was used as
a negative control. These data indicate that the S884Y mutant was also
able to display selective interaction with ER isoforms in a cellular
content.
|
and ERß and ERß was further evaluated in the LXXLL
motif peptide binding assay utilizing synthetic TRBP peptides
(878–896) mimicking the different mutations at the S884 position. In
this time-resolved fluorescence assay, synthetic TRBP LXXLL peptides
are fluorescent labeled and then assessed for their ability to bind to
recombinant ER and ERß. Consistent with our previous data, the
S884Y mutation significantly increased the interaction with ERß but
not ER (Fig. 6). A negative charge at
S884 (S884E) abolished binding of TRBP to both ER and ERß. A
positive charge at the same position (S884R) did not decrease, but
rather increased, TRBP peptide binding to ER. This effect was
particularly apparent for ER. As expected, a peptide harboring
alanine mutations (LA) of the conserved leucines was not able to bind.
In agreement with previous binding assays using large TRBP fragments,
these findings using synthetic peptides suggest that the S884 of TRBP
is an important residue responsible for the selective interactions with
ER isoforms. The binding activity within the peptide-binding assay may
not be entirely consistent with the activity of the large TRBP protein
fragments. In fact, variations of efficacy and affinity profiles for
each receptor were observed using the various assays. However, S884 was
consistently demonstrated, both in in vitro and in
vivo assays, to play an important role in ER vs.
ERß selectivity. These results indicate that the conformation of the
side chain of S884 may contribute to the binding affinities of TRBP for
ER and ERß.
|
-mediated
transcription (Fig. 7, A and B).
Consistent with our data suggesting that this mutation increased
ligand-independent interaction with ERß, the ligand-independent
transactivation mediated by ERß was also increased. In the case of
TRß, both thyroid hormone response element (F2)-reporter
activities mediated by TRß1 and Gal-reporter activity mediated by the
Gal4-TRß-LBD fusion protein were tested. In both cases, the
ligand-dependent activity was increased with the TRBP S884Y mutant
(Fig. 7, C and D). Taken together, these results show that S884 affects
the selectivity of TRBP for nuclear receptor interaction, as well as
its coactivation potential.
|
at Glu380, ERß,
and TRß with the TRBP LXXLL peptide, we obtained further structural
evidence for the role of the S884 position in determining the
interaction with the nuclear receptor LBD. The structure of the TRBP
LXXLL (882–895) motif was engineered based on the PPAR helix 12
(Fig. 2B). The engineered TRBP peptide docking was carried out by
three-dimensional alignment with the TRBP peptide and the GRIP-1
peptide cocrystallized with ER (36). The backbone and
the leucine residues within the LXXLL motif of GRIP-1 were aligned with
the backbone and the leucines within the TRBP LXXLL peptide. Side chain
orientations were adapted from PPAR helix 12 (40). As a
result, the L887, L890, and L891 in TRBP LXXLL motif were found to
insert into the hydrophobic coactivator-binding pocket of ER between
the charged clamp flanked by the conserved K362 in helix 3 and E542 in
helix 12 of ER (Fig. 8A). The L886 at
the -1 position of the TRBP LXXLL motif also directly contacted the
hydrophobic regions on the ER surface. Importantly, the structure
revealed that Ser884 of TRBP points toward the receptor as it had a
similar orientation as the leucine residues (Fig. 2B). In addition,
TRBP S884 may have close contact with E380 in helix 4 of ER. This
indicated that E380 of ER might influence TRBP binding via the S884
residue. Interestingly, the surfaces around S884 in ERß and TRß are
much different than in ER (Fig. 8, B and C) (38, 39).
There are more open spaces and the environment has more hydrophobic
character. This may explain why the S884Y mutation resulted in a higher
binding activity of the TRBP peptide for TRß and ERß than for
ER. Based on the model, it was also understandable why the
negatively charged glutamic acid residue, but not positively charged
arginine residue, at the TRBP S884 position completely abolished
binding for both ER and ERß (Fig. 6). Furthermore, when E380 of
ER was aligned with a number of nuclear receptors, amino acids at
the E380 position of ER showed conservation within nuclear receptor
subclasses (Fig. 8C). Class I receptors including AR, GR, MR, and PR
have glutamines (Q) at this position while ER and ERß have
glutamic acids (E), and class II receptors have lysine (K) or arginines
(R). It is also noteworthy that the hydrophobic character of the
coactivator binding pocket is conserved except for this charged residue
within helix 4. This indicated that each class of receptor might share
a degree of conformational similarity around this position, and that
different coactivators might be able to distinguish among various
receptor subclasses.
|
at E380 residue was
generated with the substitution of a glycine residue (E380G). The ER
E380G was compared with the wild-type ER and ERß for its binding
with a number of TRBP S884 mutants (714–1242), in which the S884 is
changed to a variety of residues including aromatic, positive, and
negative charged residues (S884E, S884F, S884H, S884K, S884L, S884P,
S884Q, S884W, S884Y). As shown in Fig. 9, the S884Y and S884F mutants had increased ERß binding, but decreased
ER binding relative to wild-type (S884), suggesting that the
aromatic side chains of phenylalanine (F) and tyrosine (Y) are
responsible for this selectivity. A mutant bearing tryptophan (W),
which is also aromatic but maybe too bulky in its side chain structure,
bound poorly to both ER and ERß. However, when the interaction of
the ER E380G mutant with the TRBP S884 mutants was compared, all
three aromatic residues including tryptophan at S884 resulted in a
dramatic increase in their binding activities, indicating that the
deletion of the E380 side chain was primarily responsible for the
increased interactions of the aromatic residues. These results strongly
suggest that the 884 in TRBP and 380 in ER may be in close contact.
Hydrophobic leucine (L) and acidic glutamine (Q) at 884 had much less
binding activity than the aromatic residue mutants (Fig. 9). Positively
charged histidine (H) and lysine (K) in that position in TRBP results
in different binding activities to ER and ERß. However, E380G of
ER did not improve binding of either, indicating that the aromatic
structure, rather than charge, may play a major role. A proline (P) at
884 completely abolished the binding. Inasmuch as the 885 position of
TRBP is already a proline in the wild-type protein, another proline
immediately next to it may produce a TRBP peptide that is too rigid to
accommodate an induced fit binding. As expected, the glutamate mutation
(S884E) abolished binding, most likely due to the proximity of other
negatively charged residues nearby on the surface of the LBD. The
precise contact of TRBP S884 with receptors remains to be established.
However, it is clear that each receptor has a unique conformation
surrounding the coactivator binding pocket. The S884 of TRBP
undoubtedly influences the binding selectivity for each of the
receptors.
|
). Alanine mutation at S884
significantly reduced, but did not completely abolish, the
phosphorylation, indicating S884 is a major phosphorylation site
for MAPK within this 137-amino acid LXXLL-containing fragment of
TRBP. To evaluate the binding by a phosphorylated TRBP to ER and
ERß, a synthetic phosphopeptide bearing a phos-S884 was tested. The
results suggested that phosphorylation at S884 of TRBP dramatically
reduced ligand-dependent TRBP interactions with both ER and ERß
when compared with the wild-type unphosphorylated form, although there
was a significant increase of interactions with the unliganded receptor
(Fig. 10B). It is currently unknown whether S884 can be phosphorylated
in vivo. However, gene-specific transcriptional
activation, at least in part, may be regulated by
phosphorylation. The phosphorylation at S884 of TRBP is able to
modulate receptor binding in vitro, suggesting the possible
regulatory interactions between coactivator and nuclear receptors, if
such phosphorylation events occur in vivo.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. In addition, we show that the S884Y mutant not only
differentially interacted with ERß vs. ER both in
vitro and in vivo, but also displayed functional
differences in ER-mediated transactivation in cells.
The interaction of the LXXLL motif with the nuclear receptor LBD is
well documented. Early mutagenesis studies established the absolute
requirement of the three leucines of the LXXLL motif as well as an
intact AF-2 domain for functional interaction of the coactivator with
the nuclear receptor (1, 9, 10). Subsequent detailed
analysis, including peptide library screens of the LXXLL motif,
revealed that each nuclear receptor has a distinct preference for the
amino acid sequences adjacent to the motif (42, 43, 44, 45, 46, 47).
Interestingly, many LXXLL peptides with hydrophobic residues at the -1
position, resulting in a 
LXXLL motif, had higher binding affinity to
nuclear receptors such as TR and ERß (44, 45). This
character is also found in the majority of naturally existing
LXXLL-containing coactivators, which includes most of the motifs of the
three members of the SRC-1 family, both motifs of TRAP220, a number of
motifs in RIP140, and the single motif in both PGC-1 and TRBP.
More interestingly, a proline at the -2 position (PLXXLL) confers
the highest selectivity for TRß in a peptide library screen
(45). This PLXXLL motif is found naturally in both TRBP
and TRAP220 and, interestingly, both of these coactivators were
initially identified with high affinity for TR.
Several points were considered for the rationale to identify the amino
acids near the LXXLL motif for residues that may be important for
receptor selectivity. First, small synthetic coactivator peptides can
sometimes yield high binding affinity, but may not reflect
physiological situations in which binding may be regulated and optimal
for in vivo function. We used a natural coactivator TRBP
fragment (528 amino acids) to overcome this limitation. Our data also
suggested that an even longer TRBP fragment (1,350 amino acids, Fig. 7)
was also able to retain the selectivity. Second, amino acid residues
immediately adjacent to the LXXLL motif were selected for mutation
since they are likely to have close contact with the receptor and thus
may be important for selectivity. Since the LXXLL motif is -helical
in structure, the -3, -4, +7, and +8 positions are considered to be
the most immediate residues next to the conserved leucines that are
likely to contact the LBD surface. Because a PCR method can be used for
convenient introduction of mutations, we undertook this approach in
which mutations were randomly generated only at these four positions in
a 528-amino acid TRBP fragment containing the LXXLL motif. Among 64
(43) possible nucleotide combinations for each
amino acid, three will be stop codons (TAA, TGA, TAG). In theory,
17.4% (1 - (64 -
3)4)/644) of the
mutations will be stop codons, and a minimum of 22 would need to be
screened to have each residue appear at least once, by chance, at each
of the four positions.
It was expected that most mutations near the motif would reduce the
binding to the receptor since it was most likely that alteration of the
natural structure of the protein would result in deleterious effects on
function. The results did confirm that the majority of the clones
displayed less binding than wild type (Fig. 1). However, the limited
mutants displaying increased binding provided an opportunity to
identify the residues that may be important. Among these mutants,
clones 6 and 33 had a significant increase in binding to TR and RXR.
Notably, these two clones also selectively favored interaction with
ERß over ER. Comparison of these clones suggested that aromatic
residues at the -3 position might correlate with the ERß
vs. ER selectivity, and positive residues at the -4
position (clone 12 and 14) might display selectivity for TRß
vs. RXR. On the other hand, hydrophobic residues at +8
position in clone 11,19, 25, 26, and 27 almost abolished TR, RXR, and
ERß binding, but not ER binding. These data indicated that the
structural determinants on the surface of TR, RXR, and ERß might
share similarities that differ from ER.
Crystallographic studies with several nuclear receptors suggested that
most of the residues comprising the coactivator-binding groove are
conserved among receptors. They are nonpolar except for two highly
conserved charged residues, which form the charge clamp. A glutamic
acid residue within helix 12 and an lysine residue in helix 3 have been
demonstrated to be indispensable for LXXLL peptide binding since
mutations at either of these sites abolishes coactivator-peptide
interactions (50). Our modeling studies revealed a residue
near this region, corresponding to E380 in ER, while variable among
members of the entire nuclear receptor superfamily, is conserved within
receptor subfamilies. The steroid receptors, including AR, GR, MR, and
PR, contain a glutamine residue at this position, whereas ER and
ERß contain a glutamic acid residue. The class II receptors have
either lysine or arginines at this position (Fig. 8D). Because the -2
position of the TRBP LXXLL motif is a proline, which breaks the helix
and results in the side chain of S884 to be directed toward helix 4 of
the ER LBD, it was not surprising that S884 is a critical residue
when the peptide structure of the TRBP LXXLL motif was modeled with the
ER crystal structure. This structural model prompted us to predict,
and further demonstrate, that the E380 residue closely interacts with
S884 of TRBP. In the case of TRß, RXR, and ERß, this location is
relatively open and composed of mostly hydrophobic residues. This may
explain why a tyrosine substitution at the -3 position increased TRBP
binding to TRß, RXR, and ERß but not to ER. Because the side
chain of the S884 residue of TRBP is predicted to be in close proximity
to this variable site in the coactivator binding region of the
nuclear receptor, it is likely that the unique chemical character, as
well as the relative spacing displayed by the variable nuclear receptor
residue, may play a role in TRBP selectivity and discrimination between
various nuclear receptor subfamilies. Interestingly, the S884 residue
can be phosphorylated by MAPK in vitro (Fig. 10A),
and phosphorylation alters the interaction of TRBP to ER and ERß
(Fig. 10B). These data indicate that TRBP, when posttranslationally
modified, may exhibit differential interactions with various nuclear
receptors.
Characterization of the unique requirements for nuclear
receptor-specific coactivator binding may be useful for the design of
peptide antagonists that selectively block coactivator binding. This
may be useful for the development of ER isoform-selective antagonists.
The treatment and prevention of ER-related diseases and disorders among
tissues such as breast, bone, urogenital system, cardiovascular system,
and central nervous system rely on our understanding of ER-
mediated transcriptional regulation (48, 52). ER
and ERß have unique tissue distributions as well as distinct
biological functions. ER is predominantly expressed in many tissues
including the uterus, mammary gland, bone, brain, liver, heart, kidney,
and pituitary, while ERß mRNA is significant in ovary and prostate
(48, 51). Knockout studies have revealed that phenotype
differences exist between ER- and ERß-deficient animals, such as
breast tissue development (48). Consistent with the
observation that ER modulator-induced conformation changes on ER and
ERß are distinct (43, 53, 54), our work provides further
evidence that coactivators might differentially interact with the two
receptors. Thus, a better understanding of ER action, including the
selective activation of ER and ERß, may be helpful in the
development of such therapeutic agents.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ERß was purchased from Pan Vera (Madison, WI). The N-terminal
biotin-labeled TRBP peptides (878–896) with mutations at S884 were
synthesized by Sigma-Genosys (Woodlands, TX). The human
TRBP plasmid used in this study was previously described
(30). The TRBP fragment (714–1242) containing the LXXLL
motif was subcloned into pcDNA3 and used for in vitro
translation. Nuclear receptor LBDs were inserted in frame into pGEX
4T-2 (Amersham Pharmacia Biotech, Piscataway, NJ) to
produce the glutathione S-transferase (GST)-LBD fusion
proteins (TRß, RXR, ER, ER E380G, and ERß). Full-length human
TRß1, ER, and ERß were subcloned into pcDNA3
(Invitrogen, San Diego, CA). F2/TRE and 2XERE luciferase
reporters were previously described (30). The
5XGal-luciferase reporter pFR-luc was from Stratagene (La
Jolla, CA). Mammalian two-hybrid plasmid vectors pM (Gal DNA-binding)
and pVP16 were from CLONTECH Laboratories, Inc. (Palo
Alto, CA). GAL-TRBP (795–999) mutants S884Y and S884E were produced by
PCR and inserted in frame at the C terminus of GAL4 DNA-binding domain
in the pM vector. The GAL-TRBP (795–999) alanine mutant (LA) was
constructed by mutating the N889, L890, and L891 residues of TRBP to
alanine.
Random Mutagenesis of the TRBP LXXLL Motif and Binding Analysis
of Individual Clones
Mutations were randomly introduced at amino acid positions -3,
-4, +7, and +8 of the TRBP LXXLL motif
(LTSPLLVNLLQSDI) using PCR. The
underlined amino acids indicate the positions of the
residues that were randomly mutated. Two degenerate primers were
synthesized and used for PCR, producing random mutations at designated
locations within a 528-amino acid TRBP fragment (714–1242). The
primers used were P1: TGTTGGTCAACTTATTGCAGNNNNNNATATCTGCAGGCCAT; and
P2: AATAAGTTGACCAACAATGGNNNNNNTAGCGTGACATCCTT. PCR products that
contained a mixture of all of the different mutations were subcloned
into the pcDNA vector under the control of the bacterial T7 promoter.
The plasmid DNA clones for each mutation clones were separated and
isolated by plasmid minipreps, and individually in vitro
translated and [35S]methionine labeled. The
labeled TRBP proteins carrying a variety of mutations were incubated
with GST-LBD of TRß, RXR, ER and ERß in a binding assay. The
binding assays were performed at room temperature for 1 h in
binding buffer (20 mM HEPES, pH 7.4, 50
mM NaCl, 75 mM KCl, 1
mM EDTA, 0.05% Triton X-100, 10% glycerol, 1
mM dithiothreitol). Bound proteins were washed
three times with binding buffer and subjected to SDS-PAGE followed by
autoradiography. Equal amounts of the GST-LBD proteins within each
receptor group were used. The signal strength was optimized with the
exposure time of film to obtain the best contrast within each receptor
groups.
Recombinant Protein Binding Assays
The GST fusion nuclear receptor LBDs were produced in
Escherichia coli BL21(DE3) and purified with
glutathione-Sepharose resin (Amersham Pharmacia Biotech).
In vitro binding assays were performed by incubating GST-LBD
resin (15 µl, 2–4 µg) and
[35S]methionine-labeled, in vitro
translated TRBP mutant proteins (5 µl) produced by rabbit
reticulocyte lysate (Promega Corp., Madison, WI). Proteins
were incubated at room temperature for 1 h in binding buffer (20
mM HEPES, pH 7.4, 50 mM
NaCl, 75 mM KCl, 1 mM EDTA,
0.05% Triton X-100, 10% glycerol, 1 mM
dithiothreitol). Bound proteins were washed three times with binding
buffer and subjected to SDS-PAGE followed by autoradiography or
quantitation using a scintillation counter. For the phosphopeptide
binding assay, GST-LBD resins (15 µl, 2–4 µg) were incubated with
the biotin-labeled peptides (3 µM, WT:
KDVTLTSPLLVNLLQSDIS; pS884: KDVTLTpSPLLVNLLQSDIS) in the presence or
absence of E2 (1 µM) and washed as described
above. The GST-LBD beads were incubated with
streptavidin-peroxidase (1 U/ml, Roche Molecular Biochemicals, Indianapolis, IN) and washed three times. The
beads were then transferred to the 96-well plate, and bound peptides
were detected by chemiluminescence with ECL detection reagents
(Amersham Pharmacia Biotech).
Peptide Binding Assays
The details of methods used in the peptide binding assay
was as previously described (55). Briefly, white
FluoroNunc 96-well plates were coated overnight with 4 pmol/well of
baculovirous-expressed full-length human ER or ERß. The
receptor-coated plates were blocked with 7.5% BSA in blocking buffer
TBS (0.1 M Tris-HCl, 0.15 M NaCl, 20
µM diethylenetriamine-pentaacetic acid) at room
temperature for 1 h. The plates were then washed three times with
0.1% Triton X-100 in TBS (TBST). Biotin-labeled TRBP peptides (6.6
nM) carrying mutations at S884 were preincubated with
europium-conjugated streptavidin (0.83 µg/ml) on ice for 30 min.
Peptide-europium conjugates (120 µl) were diluted with DELFIA assay
buffer (Perkin-Elmer Corp., Gaithersburg, MD) into
10 ml and applied to the 96-well plate with 90 µl/well. E2 was then
added at appropriate concentrations to each well, and plates were
incubated for 1.5 h at room temperature followed by five washes
with TBST. The signal was detected by adding 100 µl of DELFIA
Enhancement Solution (Perkin-Elmer Corp.) with 5 min of
gentle shaking followed by analysis using a Wallac, Inc.
Victor II plate reader. The peptides used in this assay are as
follows. WT: KDVTLTSPLLVNLLQSDIS; S884R: KDVTLTRPLLVNLLQSDIS; S884Y:
KDVTLTYPLLVNLLQSDIS; S884E: KDVTLTEPLLVNLLQSDIS; LA:
KDVTLTEPLLVNAAQSDIS.
Cell Culture and Transient Transfection
CV-1 cells were maintained in DMEM supplemented with 10% FBS
and 0.1 µg/µl penicillin/streptomycin in 5%
CO2 at 37 C. Cells were plated in 24-well plates
for 2 d before transfection. CV-1 cells were transfected using
lipofect- AMINE reagent according to the manufacturer’s protocol
(Life Technologies, Inc., Gaithersburg, MD). Cells were
incubated in fresh serum-free medium containing 100 µM of
ligand for 16–24 h after transfection. Total amounts of DNA for each
well were balanced by adding vector DNA pcDNA3
(Invitrogen). Relative luciferase activities were measured
and shown as means of triplicate transfections ± SE
values.
Crystal Structure Modeling with TRBP Peptide
TRBP peptide structure (882–895, LTSPLLVNLLQSDI)
was computationally engineered based on the backbone of the PPAR
AF-2 helix (SLHPLLQEIYKLDL) (2PRG) (37). The
AF-2 domains were shown to contact the LXXLL-binding pocket and
structurally overlap with the LXXLL helices (1, 10, 36).
In addition, the AF-2 helices in a number of receptors align with the
coactivator LXXLL motifs (37). Since the PPAR AF-2 has
a leucine at the -1 and a proline at the -2 positions, the backbone
of PPAR AF-2 resembles the backbone of TRBP LXXLL peptide near the
-3 position. Amino acid side chains were computationally engineered
using the QUANTA program. The three hydrophobic leucines of the
engineered TRBP peptide were aligned with the three leucines of GRIP-1
peptide in the three-dimensional crystal structure of ER or TRß.
In the case of ERß, the leucines of the TRBP LXXLL were aligned with
the corresponding hydrophobic residues of helix 12 in ERß.
Coordinates of ER/DES/GRIP-1 (3ERD) (36),
ERß/genestein (1QKM), and TRß/GRIP-1 (1BSX) were obtained from the
Protein Data Bank (38, 39).
In Vitro Kinase Phosphorylation
In vitro phosphorylation of recombinant His-tagged
TRBP (795–931) by MAPK was assayed using the SigmaTECT Protein Kinase
Assay buffer system from Promega Corp. with modifications.
Briefly, MAPK (ERK2, New England Biolabs, Inc., Beverly,
MA) was incubated in supplied buffer system with
32P--ATP and His-tagged wild-type TRBP
(795–931) or TRBP S884A (795–931) as substrates. Labeled His-tagged
protein beads were washed, and the proteins were resolved by SDS-PAGE
followed by autoradiography.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: ACTR, Activator of thyroid receptor; AF-2,
activation function 2; AIB, amplified in breast cancer; ASC-2,
activating signal cointegrator-2; CBP, cAMP response element binding
protein binding protein; CIA, coactivator independent of AF-2 function;
CoAA, coactivator activator; DES, diethylstilbestrol; DRIP,
vitamin D receptor-interacting protein; GRIP, GR-interacting
protein; GST, glutathione-S-transferase; LBD,
ligand-binding domain; NcoA, nuclear receptor coactivator; NRC, nuclear
receptor cointegrator; PBP, PPAR binding protein; PGC-1, PPAR
coactivator-1; RAC-3, receptor associate coactivator 3; RRM, RNA
recognition motif; SRC, steroid receptor coactivator; TIF-2,
transcriptional intermediary factor 2; TRAP, TR associate protein;
TRBP, TR binding protein.
Received for publication May 17, 2001. Accepted for publication September 14, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. J Biol Chem 273:31108–31112 and ER ß. Proc Natl Acad Sci USA 96:3999–4004
and ß. Mol Cell Biol 19:8226–8239
and ß. Endocrinology 138:863–870 and ERß. J Biol Chem [epub ahead of
print]
This article has been cited by other articles:
 |
|
 |
|
  H. J. Dubbink, R. Hersmus, A. C. W. Pike, M. Molier, A. O. Brinkmann, G. Jenster, and J. Trapman Androgen Receptor Ligand-Binding Domain Interaction and Nuclear Receptor Specificity of FXXLF and LXXLL Motifs as Determined by L/F Swapping Mol. Endocrinol., August 1, 2006; 20(8): 1742 - 1755. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Mahajan and H. H. Samuels Nuclear Hormone Receptor Coregulator: Role in Hormone Action, Metabolism, Growth, and Development Endocr. Rev., June 1, 2005; 26(4): 583 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Liu, B. O. Kim, C. Kao, C. Jung, J. T. Dalton, and J. J. He Tip110, the Human Immunodeficiency Virus Type 1 (HIV-1) Tat-interacting Protein of 110 kDa as a Negative Regulator of Androgen Receptor (AR) Transcriptional Activation J. Biol. Chem., May 21, 2004; 279(21): 21766 - 21773. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Smith and B. W. O'Malley Coregulator Function: A Key to Understanding Tissue Specificity of Selective Receptor Modulators Endocr. Rev., February 1, 2004; 25(1): 45 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Takeuchi, H. H. Q. Heng, C. J. Ye, S.-B. Liang, J. Iwata, H. Sonobe, and Y. Ohtsuki Down-Regulation of a Novel Actin-Binding Molecule, Skeletrophin, in Malignant Melanoma Am. J. Pathol., October 1, 2003; 163(4): 1395 - 1404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-M. Leduc, J. O. Trent, J. L. Wittliff, K. S. Bramlett, S. L. Briggs, N. Y. Chirgadze, Y. Wang, T. P. Burris, and A. F. Spatola Helix-stabilized cyclic peptides as selective inhibitors of steroid receptor-coactivator interactions PNAS, September 30, 2003; 100(20): 11273 - 11278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. H. Coulthard, S. Matsuda, and D. M. Heery An Extended LXXLL Motif Sequence Determines the Nuclear Receptor Binding Specificity of TRAP220 J. Biol. Chem., March 21, 2003; 278(13): 10942 - 10951. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ko and W. W. Chin Nuclear Receptor Coactivator Thyroid Hormone Receptor-binding Protein (TRBP) Interacts with and Stimulates Its Associated DNA-dependent Protein Kinase J. Biol. Chem., March 21, 2003; 278(13): 11471 - 11479. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. He and E. M. Wilson Electrostatic Modulation in Steroid Receptor Recruitment of LXXLL and FXXLF Motifs Mol. Cell. Biol., March 15, 2003; 23(6): 2135 - 2150. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Wu, W. W. Chin, Y. Wang, and T. P. Burris Ligand and Coactivator Identity Determines the Requirement of the Charge Clamp for Coactivation of the Peroxisome Proliferator-activated Receptor gamma J. Biol. Chem., February 28, 2003; 278(10): 8637 - 8644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-H. Goo, Y. C. Sohn, D.-H. Kim, S.-W. Kim, M.-J. Kang, D.-J. Jung, E. Kwak, N. A. Barlev, S. L. Berger, V. T. Chow, et al. Activating Signal Cointegrator 2 Belongs to a Novel Steady-State Complex That Contains a Subset of Trithorax Group Proteins Mol. Cell. Biol., January 1, 2003; 23(1): 140 - 149. [Abstract] [Full Text]
|
|
|
|
|
|
T. Suzuki, M. Kasahara, H. Yoshioka, K.-i. Morohashi, and K. Umesono LXXLL-Related Motifs in Dax-1 Have Target Specificity for the Orphan Nuclear Receptors Ad4BP/SF-1 and LRH-1 Mol. Cell. Biol., January 1, 2003; 23(1): 238 - 249. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. Schiff, S. Massarweh, J. Shou, and C. K. Osborne Breast Cancer Endocrine Resistance: How Growth Factor Signaling and Estrogen Receptor Coregulators Modulate Response Clin. Cancer Res., January 1, 2003; 9(1): 447S - 454S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Malloy, R. Xu, L. Peng, P. A. Clark, and D. Feldman A Novel Mutation in Helix 12 of the Vitamin D Receptor Impairs Coactivator Interaction and Causes Hereditary 1,25-Dihydroxyvitamin D-Resistant Rickets without Alopecia Mol. Endocrinol., November 1, 2002; 16(11): 2538 - 2546. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. J. McKenna and B. W. O'Malley Minireview: Nuclear Receptor Coactivators--An Update Endocrinology, July 1, 2002; 143(7): 2461 - 2465. [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 |
