| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Metabolic Research Unit University of California San Francisco, California 94143-0540
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ERß cooperative activation with Pit-1
suggested a conserved ER site for RIP140 action, possibly AF-2.
Coexpression of other AF-2-interacting proteins, including the p160
factors, steroid receptor coactivator-1a (SRC-1a) and glucocorticoid
receptor interacting protein-1 (GRIP1), had negligible effects on
ER/Pit-1 cooperative activation, but partially relieved RIP140
inhibition. Relief of RIP140 inhibition required the AF-2-binding,
LXXLL motifs in SRC-1a and GRIP1. An ER AF-2 mutant that selectively
blocked ER interaction with p160s, but not RIP140, still cooperated
with Pit-1 and was inhibited by RIP140, but was not relieved by SRC-1a
or GRIP1 expression. Thus, SRC-1a and GRIP1 binding to AF-2
counteracted the inhibition of ER/Pit-1 activation by another
AF-2-interacting protein, RIP140. Complex, sometimes antagonistic,
actions of different classes of AF-2-interacting proteins may play an
important role in the cell- and gene-specific estrogen regulation of
PRL and other genes. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Many estrogen responses are mediated by the estrogen receptor (ER), a DNA-binding transcription factor of the nuclear receptor superfamily that generally is more effective at promoting transcription when bound by estrogens (1, 2, 3, 4, 5, 6). ER activation is associated with the estrogen-dependent interaction of certain receptor-interacting proteins (RIPs) with activation function-2 (AF-2) (7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19), which is conserved in the carboxyl-terminal ligand-binding domain (LBD) of ER and most nuclear receptors (20). A separable amino-terminal domain of ER contains the more poorly conserved AF-1, which in most promoter contexts is dormant unless released by agonist binding to the LBD (2, 4, 21, 22). Ligand dependence of AF-1 may be related to relief of corepression at many nuclear receptors (23, 24, 25, 26, 27) or may be associated with the cobinding of the same receptor interacting coactivator to AF-1 and AF-2 (28, 29, 30).
AF-1 and AF-2 have been described mostly in studies of ER action at short, isolated DNA sequences to which ER binds. Estrogen regulation of natural genes is likely to use many of the same ER activities in conjunction with the activities provided by other promoter-binding factors. Synthesis of PRL declines with age in females and is strongly regulated by estrogens. Estrogen activation of pituitary-specific PRL gene transcription provides an excellent model in which to study cell-specific modulation of ER action by other promoter-binding factors.
ER regulates PRL gene transcription only if the PRL enhancer/promoter is bound by the pituitary-specific transcription factor Pit-1 (31, 32, 33, 34, 35). This cooperative activation by ER and Pit-1 explains why estrogen regulation of PRL gene transcription is pituitary specific. ER/Pit-1 cooperative activation requires AF-2 (35). However, RIP140, a 140-kDa cofactor that interacts with AF-2 of ER and a number of nuclear receptors in a ligand-dependent fashion (18), inhibits ER/Pit-1 activation (35). RIP140 may inhibit any of the PRL enhancer/promoter-binding proteins including ER. Understanding how RIP140 and other AF-2-interacting proteins modulate AF-2-dependent ER/Pit-1 cooperative activation will be crucial for understanding cell-specific estrogen regulation of gene transcription.
In the present study, it was observed that RIP140 inhibition of
ER/Pit-1 PRL enhancer/promoter activation operated completely via ER.
Within ER, AF-1 was not required for RIP140 inhibition. RIP140
inhibition of the PRL enhancer/promoter activated by Pit-1 and either
of the - or ß-isoforms of ER suggested a conserved site for RIP140
action, possibly AF-2. Relief of RIP140 inhibition by the coexpression
of other AF-2-interacting factors supported AF-2 as the target of
RIP140 action. Disrupting interaction of those counteracting factors
with AF-2 blocked their ability to relieve RIP140 inhibition. Thus,
AF-2 dependent promoter- and cell-specific ER regulation of the PRL
enhancer arises from the cooperative and antagonistic interactions of
Pit-1, ER, and AF-2-interacting proteins docking directly to ER
AF-2.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In GHFT1–5 cells, PRL enhancer/promoter activity was low unless
an expression vector for the cDNA of Pit-1 was cotransfected (34, 35)
(Fig. 1A
). Expression of the cDNA for the
-isoform of human ER (ER) had a marginal effect on PRL
enhancer/promoter activity. In contrast, activation was substantial
when ER and Pit-1 were coexpressed, averaging 25.1 ± 5.1-fold
activation over the uninduced PRL enhancer/promoter.
|
(13), and RIP140 (18) were
individually cotransfected along with the ER and Pit-1 expression
vectors and the PRL enhancer/promoter reporter into GHFT1–5 cells
(Fig. 1A
). PRL enhancer/promoter activities in the presence of ER,
Pit-1, and SRC-1a, GRIP1, or TIF1 were 128 ± 29%, 129 ±
31%, and 129 ± 13%, respectively, as active as PRL
enhancer/promoter activity in the absence of coactivator. Similarly,
the coactivator CREB-binding protein (CBP) (38), which also
participates in nuclear receptor action (39), did not significantly
alter ER/Pit-1 activation of the PRL enhancer/promoter (120 ±
14%).
RIP140 Is a Dose-Dependent Inhibitor of ER/Pit-1 Synergy
Expression of RIP140, which, like SRC-1a, GRIP1, and TIF1,
binds AF-2 in a ligand-dependent fashion, strongly reduced ER/Pit-1
cooperative activation of the PRL enhancer/promoter (35) (Fig. 1A).
RIP140 inhibition was dose dependent; transfection of 1–10 µg of the
RIP140 expression vector caused a gradual decrease in ER/Pit-1
cooperative activation (Fig. 1B). ER/Pit-1 activation was effectively
abolished to the level of Pit-1 activation by transfecting 10 µg of
the RIP140 expression vector. In contrast, cotransfection of 10 µg of
the RIP140 expression vector had no effect on C/EBP activation of
the PRL enhancer/promoter (35). Transfection of lower amounts of the
same RIP140 expression vector was previously reported to enhance ER
activation at a minimal promoter containing an estrogen response
element (ERE) (18), but in the current studies had no activating or
inhibitory effect on ER/Pit-1 cooperative activation of the PRL
promoter (data not shown).
The levels of ER and Pit-1 expressed were determined by Western blots
of nuclear extracts of the same transfected cells (Fig. 1C).
Ectopically expressed human ER and Pit-1 (with the FLAG epitope
fused to the Pit-1 amino-terminus) were detected with antibodies
specific for human ER and the FLAG epitope. Mouse GHFT1–5 cells
contain some endogenous ER (35) that were not detected by the
anti-human ER antibodies used here. The nuclear levels of ectopically
expressed ER or Pit-1 were not changed by the coexpression of Pit-1
or ER (Fig. 1C) by RIP140 (Fig. 1C) or by SRC-1a, GRIP1, or TIF1
expression (not shown). Similarly, endogenous ER and Pit-1 levels were
not changed (35). This showed that ER/Pit-1 activation and its
regulation by RIP140 were not artifacts of effects on the expression
vectors themselves. Thus, the AF-2-interacting protein RIP140 directly
inhibited estrogen-dependent, cooperative activation by ER and
Pit-1.
ER Is Required for RIP140 Inhibition
RIP140 may inhibit the PRL enhancer/promoter by acting directly at
the ER. Conversely, RIP140 may inhibit other factors, including Pit-1,
that bind to other PRL enhancer/promoter sites (40, 41, 42, 43, 44, 45, 46). We determined
that RIP140 did not affect PRL enhancer/promoter activity in the
absence of ER (Fig. 2). Because GHFT1–5
cells contain endogenous ER (35), we conducted these experiments under
conditions in which endogenous ER was selectively inactivated by
incubating the transfected cells with 10-6 M
ICI 164,384. ICI 164,384 is an inactive estrogen mimic that antagonizes
estrogen action by competitive binding to the ligand-binding pocket of
ER. A concentration of 10-6 M ICI 164,384 was
sufficient to completely block ER/Pit-1 cooperative activation of the
PRL enhancer/promoter (data not shown).
|
). Thus, RIP140
inhibition of PRL enhancer/promoter specifically required the presence
of active ER.
RIP140 Directly Inhibits ER Action
We showed above that ER is an essential participant in RIP140
inhibition of ER/Pit-1 cooperative activation at the PRL
enhancer/promoter. To determine whether ER itself could be the site of
RIP140 action, we studied ER action in isolation from all PRL
enhancer/promoter elements. The well characterized vitellogenin ERE was
appended to a minimal promoter containing the TATA box of the herpes
simplex thymidine kinase gene (ERE-TATA). Like the PRL ERE, the
vitellogenin ERE operates through what is referred to as the classical
ER pathway (47) (see Fig. 4). The vitellogenin ERE can also replace the
PRL ERE to support ER cooperative activation with Pit-1 (33).
|
). E2 activation was
completely inhibited by 10-6 M ICI
164,384 (ICI+E2; Fig. 3). Therefore, E2
activation of the ERE-TATA promoter operated via endogenous ER
and/or ERß, both of which are detected by Western blots of nuclear
extracts of GHFT1–5 cells (35) (our unpublished data).
|
) to 5 µg (data not shown) of ER
expression vector only modestly enhanced E2 activation of
the ERE-TATA promoter 1.3-fold, on the average. In contrast,
E2 induction of the PRL enhancer/promoter depended on the
ectopic expression of ER along with Pit-1 (Fig. 4A); maximum levels of ER/Pit-1
cooperative activation of the PRL enhancer/promoter were not reached
even at 10 µg of the same ER expression vector (35). Thus,
E2 activation was saturated at much lower levels of ER at
the ERE-TATA promoter than at the PRL enhancer/promoter. The
differential response of different promoters to cellular ER levels may
represent an overlooked component of the cell- and promoter-specific
actions of ER.
Cotransfection of the RIP140 expression vector reduced E2
activation of the vitellogenin ERE-TATA promoter (Fig. 3). In the
presence of 0.2 µg of cotransfected ER, RIP140 reduced
E2-dependent ERE-TATA promoter activity to 60.8 ±
7.6%. Therefore, RIP140 inhibition was not limited to the PRL
enhancer/promoter and even occurred at a minimal promoter at which the
effects of promoter binding factors other than ER were minimized.
RIP140 is likely to directly inhibit ER action at a variety of EREs in
a variety of enhancer/promoter contexts.
AF-2 Dominates ER/Pit-1 Activation of the PRL Enhancer/Promoter
The strong inhibition of ER/Pit-1 cooperative activation of the
PRL enhancer/promoter by RIP140 might reflect RIP140 inhibition via ER
AF-2 to which RIP140 is known to bind. Conversely, RIP140 might target
a site in ER other than or in addition to AF-2. A second ER activation
function, AF-1, is located in a domain separable from the LBD in which
AF-2 resides. Some AF-2-interacting proteins have been observed to
interact also with AF-1 of some nuclear receptors, including ER (28, 29). In the context of the full-length ER, AF-1 is dormant until
released by ligand binding to the LBD. To study which domains of ER
were required for RIP140 inhibition, we initially distinguished the
relative contributions of AF-1 and AF-2 to ER/Pit-1 activation of the
PRL enhancer/promoter pharmacologically.
GHFT1–5 cells were grown in medium containing low levels of estrogens
and then transfected with ER and/or Pit-1 expression vectors.
Incubation with 10-8 M E2
resulted in significant activation of the PRL enhancer/promoter by ER
and Pit-1 (Fig. 4A). E2 had little effect on the PRL
enhancer/promoter activated only by Pit-1 expression (Fig. 4A). The
small, but reproducible, activation observed with ER/Pit-1 expression
in the absence of any ligand was due to residual estrogens in the
medium, as it was blocked by 10-6 M ICI
164,384, which prevents both AF-1 and AF-2 activities.
Other ligands, including tamoxifen and raloxifene, that bind the E2-binding pocket of ER, commonly activate AF-1 (22, 28, 47, 48) without allowing the ER LBD to adopt the E2-dependent conformation in which AF-2 is available for coactivator binding (49). Tamoxifen and raloxifene are thereby partial agonists at promoters in which AF-1 is sufficient for activation. Incubation of the transfected cells with 10-6 M tamoxifen or 10-6 M raloxifene did not activate PRL enhancer/promoter activity. This suggested that ER/Pit-1 activation of the PRL enhancer/promoter is relatively AF-2 driven, with little or no contribution of AF-1 in the absence of AF-2.
AF-1 Is Not Required for RIP140 Inhibition of ER/Pit-1 Cooperative
Activation
The lack of ER/Pit-1 activation in the presence of tamoxifen and
raloxifene suggested that active AF-1 alone was insufficient for
estrogen regulation of the PRL enhancer/promoter. However, AF-2 was
also inhibited by these ligands. To examine whether any portion of
ER/Pit-1 cooperative activation was affected by AF-1, we studied the
ability of an ER deleted of its entire amino-terminal domain, including
AF-1 (ER-
N), to activate the PRL enhancer/promoter. ER-N was as
effective as full-length ER in its ability to cooperate with Pit-1 and
activate the PRL enhancer/promoter (Fig. 5). Thus, AF-1 makes no contribution to
ER/Pit-1 activation. Moreover, RIP140 also inhibited ER-N/Pit-1
activation as efficiently as ER/Pit-1 activation (Fig. 5). Therefore,
despite the contribution of some AF-2-interacting proteins to AF-1/AF-2
cooperation (28, 29), RIP140 inhibition is complete in the absence of
AF-1.
|
knockout mice is reduced,
but not eliminated (52), suggesting a role for ER in PRL gene
regulation that might be partially compensated by pituitary ERß. Like
ER (Fig. 4A), ERß activation of the PRL enhancer/promoter in
GHFT1–5 cells grown in E2-deficient medium depended upon
Pit-1 and E2 (Fig. 4B).
Estrogen regulation results from classical ER interactions with an ERE
in the PRL enhancer. Estrogen regulation also may arise from ER
interactions with other factors, including AP-1 (47), which is a potent
regulator of the PRL promoter (44, 53). We used differences in the
pharmacology of the estrogen response via the classical and AP-1
pathways to determine their respective contributions to Pit-1
cooperative activation with both ER and ERß. In general, ER and
ERß activation via AP-1 are mechanistically distinct and
distinguishable by differing agonist/antagonist profiles (48). At AP-1,
tamoxifen acts as an agonist for both isoforms, and raloxifene
selectively activates ERß. Both ER/Pit-1 and ERß/Pit-1
activation of the PRL enhancer/promoter were insensitive to tamoxifen,
raloxifene, and ICI 164,384 and were only activated by estradiol (Fig. 4). The lack of any agonist activity by either tamoxifen or raloxifene
suggested that the AP-1 pathway did not contribute to estrogen
regulation of Pit-1 cooperation with either ER isoform.
RIP140 Inhibits Pit-1 Cooperation with ERß
The similar pharmacology and similar Pit-1 dependence of ER and
ERß suggest that the two ER isoforms play similar roles in the
estrogen regulation of the PRL gene. ER and ERß are homologous
over the DNA-binding domain and LBD (50), which are sufficient
for RIP140 inhibition of ER/Pit-1 cooperative action (Fig. 5). We,
therefore, examined whether RIP140 would inhibit ERß cooperative
activation with Pit-1.
Like ER, the coexpression of RIP140 completely blocked ERß/Pit-1
cooperative activation (Fig. 6). RIP140
expression also did not change the relative abilities and inabilities
of E2, tamoxifen, raloxifene, or ICI 164,384 to promote
ER or ERß cooperative activation with Pit-1 at the PRL
enhancer/promoter (data not shown). Thus, ER and ERß behave
similarly with respect to RIP140 inhibition of ER/Pit-1 cooperativity
at the PRL enhancer/promoter.
|
and ERß action at the PRL
enhancer/promoter most likely indicated that they possess a common
target for RIP140 action. Both ER and ERß contain AF-2, which is
required for direct RIP140 binding to ER and other nuclear receptors.
Within AF-2, the binding sites for some of the AF-2-interacting
proteins overlap, albeit with AF-2 amino acid requirements specific for
each interacting protein (8, 16, 54, 55, 56, 57). Consistent with binding site
overlap, at least RIP140 and SRC-1a have been shown to compete for
binding to PPAR
(58).
If AF-2 were a target for RIP140 inhibition, other AF-2-interacting
proteins that did not inhibit ER/Pit-1 cooperative interaction (Fig. 1)
should compete with RIP140 binding and reduce RIP140 inhibition of the
PRL enhancer/promoter. Expression of SRC-1a, GRIP1, and TIF1 shifted
the dose response of RIP140 inhibition to the right (Fig. 7A). Western blots showed that these
effects were not due to changes in the expression of ER or Pit-1 (Fig. 7B). PRL enhancer/promoter activity progressively declined with higher
RIP140 amounts in the presence of each coactivator (Fig. 7A). However,
there were some differences in the coactivator responses.
|
became less effective and was
eventually eliminated at higher RIP140 expression levels. Intersection
of the RIP140 dose-response curves in the presence and absence or
TIF1 at high RIP140 levels would be expected if RIP140 competed for
and occluded TIF1 binding to the same site. At high RIP140 levels,
the RIP140 dose-response curve in the presence of GRIP1 also approached
that in the absence of GRIP1, although it did not intersect it. The
difference in the degree to which GRIP1 and TIF1 counteracted RIP140
inhibition may reflect relative differences in TIF1 and GRIP1
expression levels, interaction with AF-2, or ability to coactivate the
PRL enhancer/promoter.
In contrast, the slope of the RIP140 dose-response curve in the
presence of coexpressed SRC-1a diverged from the RIP140 dose-response
curve. This would not be anticipated if the only effect of SRC-1a was
to occlude AF-2 and prevent RIP140 from binding. This may suggest that
the SRC-1a-containing complex was a strong coactivator of ER/Pit-1
action at the PRL enhancer/promoter or that the SRC-1a-containing
complex was more stable to RIP140 competition than its GRIP1 or
TIF1-containing counterparts. An alternative possibility was that
SRC-1a acted through a different site on the ER than did GRIP1 or
TIF1.
Relief of RIP140 Inhibition Requires SRC-1a and GRIP1 Interaction
with AF-2
Inhibition of ER/Pit-1 cooperative activation of the PRL
enhancer/promoter by the AF-2-interacting protein RIP140 was therefore
counteracted by the expression of AF-2-binding coactivators. To
determine whether AF-2 itself was the site of action for either or both
SRC-1a and GRIP1, mutations that disrupt SRC-1a and GRIP1 interaction
with ER AF-2 were examined for their effect on the relief of RIP140
inhibition. SRC-1a and GRIP1 bind to AF-2 via a conserved sequence
motif, LXXLL (59), present in multiple copies in SRC-1a and GRIP1.
Point mutants changing all four copies of LXXLL in SRC-1a or two of the
three copies of LXXLL in GRIP1 to LXXAA block interaction of those
coactivators with ER AF-2 (57, 59). SRC-1a or GRIP1 containing those
same mutants were no longer capable of relieving RIP140 inhibition of
the PRL enhancer/promoter (Fig. 8). Upon
coexpression of RIP140 and mutant SRC-1a or GRIP1, PRL
enhancer/promoter activity remained at 38.8 ± 18.3% or 34.4
± 16.8% that of the ER/Pit-1-activated PRL enhancer/promoter. This
was not statistically different (P > 0.10) from the
41.8 ± 13.0% activity upon coexpression of RIP140. In contrast,
a statistically significant relief of RIP140 inhibition was observed
with the coexpression of either wild-type SRC-1a (P =
0.01) or GRIP1 (P = 0.03) to levels that were not
significantly different (P > 0.10) from those of the
enhancer/promoter activity in the absence of RIP140 or coactivator.
Thus, SRC-1a and GRIP1 relieve RIP140 inhibition through a target that
binds the LXXLL motifs.
|
). This activity was still
significantly above (P < 0.05) the level of PRL
enhancer/promoter activity that would have been obtained by the
addition of the independent Pit-1 and ER K366A activation levels.
|
). In the presence of RIP140 and SRC-1a or
GRIP1, ER-K366A/Pit-1-activated PRL enhancer/promoter activity was not
statistically different from and even tended to be lower than the
activity in the absence of coactivator. Thus, cooperative activation by
Pit-1 and the K366A mutant mouse ER was still inhibited by RIP140, but
was no longer relieved by SRC-1a or GRIP1 coexpression. Together, the
results shown in Figs. 8 and 9 conclusively demonstrated that
counteraction of RIP140 inhibition of ER/Pit-1 cooperative activation
of the PRL enhancer/promoter by SRC-1 and GRIP1 required the
interaction of both coactivators with ER AF-2. |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
ER contains at least two transcriptional activation functions, AF-1 and
AF-2, which regulate basal promoter activity (3, 4). Deletion of AF-1
had no effect on ER/Pit-1 cooperative activation at the PRL
enhancer/promoter (Fig. 5). Consistent with this, the thyroid hormone
receptor, which contains no or poor AF-1 activity (60), also cooperates
with Pit-1 to activate the pituitary-specific GH gene promoter (61). In
contrast, null mutations within ER AF-2 (35) or blocking ER AF-2
activity with the selective AF-2 antagonists tamoxifen or raloxifene
(Fig. 4) did not allow ER cooperative activation with Pit-1. Similarly,
thyroid hormone receptor cooperative activation with Pit-1 required
AF-2 (35). Thus, Pit-1 cooperative activation with at least two nuclear
receptors seems to be relatively AF-2 driven. Selective use of AF-2 may
be related to activities provided by Pit-1 that supercede, mimic, or,
in ER, preclude the use of AF-1.
Part of AF-2 consists of a conserved sequence (20) that in all nuclear
receptors forms an amphipathic -helix on one side of a
ligand-induced hydrophobic pocket. This pocket acts as a docking site
(62) for certain ER-interacting proteins, including SRC-1a, GRIP1,
TIF1, and RIP140. In mammalian cells, at least SRC-1a, GRIP1, and
RIP140 can potentiate the activities of artificial promoters in which
an ERE is placed in close proximity to the transcription start (18, 56, 57, 59) and are thus bona fide ER coactivators. However,
expression of GRIP1, SRC-1a, or TIF1 did not substantially affect
ER/Pit-1 activation of the PRL enhancer/promoter (Fig. 1). Perhaps
GRIP1, SRC-1a and TIF1 are already present in excess in GHFT1–5
cells, or Pit-1 cooperative activation with ER mutated such that it
cannot bind SRC-1 (Fig. 9A) might indicate that other known (7, 8, 9, 11, 12, 14, 17, 19, 23) or unknown AF-2-interacting coactivators
participate in ER action at the PRL enhancer/promoter.
The putative coactivator is not RIP140, which strongly inhibited
ER/Pit-1 cooperative activation at the PRL enhancer/promoter (35) (Fig. 1). RIP140 directly repressed ER in GHFT1–5 cells (Fig. 3), but did
not inhibit the PRL enhancer/promoter when ER was selectively
inactivated (Fig. 2). Thus, RIP140 inhibition of PRL gene transcription
operated via ER and not via any of the other transcription factors that
bind to the PRL enhancer/promoter. Within ER, the target of RIP140
inhibition appeared to be common to both ER isoforms (Figs. 4 and 6)
and was not affected by the deletion of AF-1.
Competition by TIF1 for RIP140 binding to a common site in ER seemed
to explain partial relief of RIP140 inhibition by TIF1 and probably
GRIP1 and SRC-1a (Fig. 7). In support of direct competition at AF-2,
the relief of RIP140 inhibition by both SRC-1a and GRIP1 was dependent
on AF-2 (Figs. 8 and 9). Similarly, RIP140 inhibition of ER/Pit-1
cooperative activation of the PRL enhancer/promoter in GHFT1–5 cells
may arise from RIP140 competitive inhibition of the binding of the
putative endogenous AF-2-interacting coactivator. The prevention of
activation by mass action interference of the binding of a strong
coactivator by the binding of a poorer coactivator, such as RIP140, is
emerging as a mechanism of corepression that is distinct from the
active mechanisms involving the recruitment of enzymatic activities.
Mass action competition at AF-2 is, however, likely to be modified by
the recruitment of activating or inhibitory complexes by the competing
factors, which may explain the observed differences in coactivator
relief of RIP140 inhibition (Fig. 7A).
Therefore, in pituitary cells, estrogen regulation of the PRL
enhancer/promoter is a synthesis of mutual, complementary, and/or
counteracting activities possessed by ER, its AF-2 interacting
cofactors, and Pit-1. The estrogen responses of other promoters might
similarly be regulated by the amount and binding preferences of
competing AF-2-interacting proteins in a cell. This would also be
influenced by the amount of ER and the differential sensitivity of some
promoters to ER levels (Figs. 3 and 4). Thus, the relative effects and
amounts of ER, different AF-2-interacting proteins, and cooperating
transcription factors will probably determine the level of
cell-specific estrogen response at many natural promoters. The
development of ER ligands that specifically affect any of those
interactions may one day provide a means to selectively activate or
inhibit estrogen responses in specific tissues.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
, 2, and 5–7). The agonist/antagonist response was determined
in cells that had been grown for 24 h before transfection in
medium lacking phenol red and containing 10% iron-supplemented newborn
calf serum tested for reduced amount of estrogens (Figs. 3 and 4).
Transfected cells were grown in the estrogen-deficient medium and
treated with 10-8 M estradiol,
10-6 M ICI 164,384, 10-6
M tamoxifen, 10-6 M raloxifene,
and/or delivery vehicles 1 day after transfection, then collected the
following day. Transfection was performed by electroporation using the
conditions and buffers previously described (45).
Collected cells were lysed in reporter lysis buffer (Promega Corp., Madison, WI). Luciferase (Figs. 1, 2, and 4–7) and
chloramphenicol acetyltransferase (Fig. 3) activities in these extracts
were determined as previously described (34, 35, 45, 61). Data from
multiple independent experiments were normalized to specific reference
points (see figure legends for n and points), and the mean ±
SD were determined for each.
For Western blots (Figs. 1 and 7), the cell pellet posttreatment with
reporter lysis buffer was resuspended in 50 mM Tris
buffered to pH 7.8 with 2-[N-morpholino]ethanesulfonic
acid, 1 mM dithiothreitol, and 0.1% Triton X-100;
pelleted; resuspended in the same buffer; then pelleted again. The
resulting crude nuclei were resuspended three times with 20 µl 20
mM HEPES (pH 7.9), 300 mM KCl, 200
mM NaCl, 1 mM EDTA, 0.1% Nonidet P-40, and
15% glycerol, and the extracts were pooled. Equivalent amounts of
extract protein (5–20 µg depending on the experiment) were loaded
onto 10% SDS-polyacrylamide gels and probed with either the FLAG M2
mouse monoclonal antibody (ICI-Kodak, Rochester, NY) or the
human-specific ER HC-20 rabbit polyclonal antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). After washing away the
primary antibodies, horseradish peroxidase-linked antimouse IgG
(Life Technologies, Gaithersburg, MD) or antirabbit IgG
(Amersham, Arlington Heights, IL) secondary antibodies
were incubated with the filters, washed, and detected with enhanced
chemiluminescence reagents (Amersham).
Plasmids
Levels of 2.5 µg (Figs. 1, 2, and 4–7) and 2.0 µg (Figs. 8 and 9) of the 3-kb PRL enhancer/promoter cloned in front of the
luciferase gene (63) were transfected in each experiment shown. The
plasmid containing the vitellogenin ERE inserted in front of the
thymidine kinase TATA box controlling the bacterial chloramphenicol
acetyltransferase gene was previously described (47); 1 µg of this
reporter was transfected in Fig. 3.
Ten micrograms of the human ER (64), 10 µg rat ERß (48), 5 µg
FLAG-tagged rat Pit-1 (35), and 1–10 µg RIP140 (18) expression
vectors were transfected into GHFT1–5 cells (Figs. 1, 2, and 4–7) as
previously described (35). Expression vectors were under the control of
the simian virus-40 (human ER), cytomegalovirus (ERß, Pit-1), and
elongation factor-1 (RIP140) promoters. A level of 0.2 µg human
ER expression vector was transfected in Fig. 3; 0.3–3 µg
wild-type or K366A mutant of the mouse ER (56) were transfected in Fig. 9. Activation of the PRL enhancer/promoter saturated at 1 µg mouse ER
expression vector, whereas saturation was not reached even with 15 µg
human ER expression vector. This presumably reflected a much higher
transcriptional activity in GHFT1–5 cells of the adenovirus major late
promoter used to express the mouse ER.
Expression vectors containing CBP (38) under the control of the Rous
sarcoma virus promoter (Fig. 1), wild-type SRC-1a (10) under the
control of the cytomegalovirus promoter (Figs. 1 and 7), wild-type
SRC-1a and SRC-1a mutated in its four LXXLL motifs (L636A/L637A,
L693A/L694A, L752A/L753A, and L1438A/L1439A; Figs. 8 and 9), wild-type
GRIP1 (15), GRIP1 mutated in its two LXXLL motifs (L693A/L694A and
L748A/L749A; Figs. 1 and 7–9) (57), and TIF1 (Figs. 1 and 7) (13),
all under the control of the simian virus-40 promoter, were previously
described. For points in which a particular cDNA was not expressed in a
matched experiment, the "empty" expression vector not containing
the specific cDNA was cotransfected.
|
ACKNOWLEDGMENTS |
|---|
); Drs. R. Kwok and R. Goodman (CBP);
and K. Paech and Dr. P. Webb (ERß). |
FOOTNOTES |
|---|
This work was supported by Grant RPG-94–028-TBE from the American Cancer Society (to F.S.).
Received for publication September 30, 1998. Revision received March 15, 1999. Accepted for publication March 19, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ERß at AP1 sites. Science 277:1508–1510 and ß. Endocrinology 138:863–870This article has been cited by other articles:
 |
|
 |
|
  A. Jean, A. Gutierrez-Hartmann, and D. L. Duval A Pit-1 Threonine 220 Phosphomimic Reduces Binding to Monomeric DNA Sites to Inhibit Ras and Estrogen Stimulation of the Prolactin Gene Promoter Mol. Endocrinol., January 1, 2010; 24(1): 91 - 103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Nagayama, S. Sasaki, A. Matsushita, K. Ohba, H. Iwaki, H. Matsunaga, S. Suzuki, H. Misawa, K. Ishizuka, Y. Oki, et al. Inhibition of GATA2-dependent transactivation of the TSH{beta} gene by ligand-bound estrogen receptor {alpha} J. Endocrinol., October 1, 2008; 199(1): 113 - 125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Ben-Jonathan, C. R. LaPensee, and E. W. LaPensee What Can We Learn from Rodents about Prolactin in Humans? Endocr. Rev., February 1, 2008; 29(1): 1 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Liu, B. Wu, J. Szary, E. M. Kofoed, and F. Schaufele Functional Sequestration of Transcription Factor Activity by Repetitive DNA J. Biol. Chem., July 20, 2007; 282(29): 20868 - 20876. [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]
|
|
|
|
|
|
D. L. Duval, M. D. Jonsen, S. E. Diamond, P. Murapa, A. Jean, and A. Gutierrez-Hartmann Differential Utilization of Transcription Activation Subdomains by Distinct Coactivators Regulates Pit-1 Basal and Ras Responsiveness Mol. Endocrinol., January 1, 2007; 21(1): 172 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Qiu, M. A. Bosch, K. Jamali, C. Xue, M. J. Kelly, and O. K. Ronnekleiv Estrogen Upregulates T-Type Calcium Channels in the Hypothalamus and Pituitary J. Neurosci., October 25, 2006; 26(43): 11072 - 11082. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Y. Lin, E. M. Resnick, and M. A. Shupnik Truncated Estrogen Receptor Product-1 Stimulates Estrogen Receptor {alpha} Transcriptional Activity by Titration of Repressor Proteins J. Biol. Chem., October 3, 2003; 278(40): 38125 - 38131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Teyssier, K. Belguise, F. Galtier, V. Cavailles, and D. Chalbos Receptor-Interacting Protein 140 Binds c-Jun and Inhibits Estradiol-Induced Activator Protein-1 Activity by Reversing Glucocorticoid Receptor-Interacting Protein 1 Effect Mol. Endocrinol., February 1, 2003; 17(2): 287 - 299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. V. Weatherman, C.-Y. Chang, N. J. Clegg, D. C. Carroll, R. N. Day, J. D. Baxter, D. P. McDonnell, T. S. Scanlan, and F. Schaufele Ligand-Selective Interactions of ER Detected in Living Cells by Fluorescence Resonance Energy Transfer Mol. Endocrinol., March 1, 2002; 16(3): 487 - 496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Lee, W. R. Duan, M. Jakacka, B. D. Gehm, and J. L. Jameson Dominant Negative ER Induces Apoptosis in GH4 Pituitary Lactotrope Cells and Inhibits Tumor Growth in Nude Mice Endocrinology, September 1, 2001; 142(9): 3756 - 3763. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Andersen and M. G. Rosenfeld POU Domain Factors in the Neuroendocrine System: Lessons from Developmental Biology Provide Insights into Human Disease Endocr. Rev., February 1, 2001; 22(1): 2 - 35. [Abstract] [Full Text]
|
|
|
|
|
|
F. Schaufele, C.-y. Chang, W. Liu, J. D. Baxter, S. K. Nordeen, Y. Wan, R. N. Day, and D. P. McDonnell Temporally Distinct and Ligand-Specific Recruitment of Nuclear Receptor-Interacting Peptides and Cofactors to Subnuclear Domains Containing the Estrogen Receptor Mol. Endocrinol., December 1, 2000; 14(12): 2024 - 2039. [Abstract] [Full Text]
|
|
|
|
|
|
C. Ying and D.-H. Lin Estrogen-modulated Estrogen Receptor{middle dot}Pit-1 Protein Complex Formation and Prolactin Gene Activation Require Novel Protein Synthesis J. Biol. Chem., May 12, 2000; 275(20): 15407 - 15412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Resnick, D. A. Schreihofer, A. Periasamy, and M. A. Shupnik Truncated Estrogen Receptor Product-1 Suppresses Estrogen Receptor Transactivation by Dimerization with Estrogen Receptors alpha and beta J. Biol. Chem., March 15, 2000; 275(10): 7158 - 7166. [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 |
