| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Molecular and Cellular Biology Baylor College of Medicine Houston, Texas 77030-3498
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
nor C/EBP
was able to substitute for
C/EBPß when cotransfected with STAT5 and GR. Different GR
determinants were needed for transcriptional cooperation between STAT5
and GR as compared with those required for all three transcription
factors. These studies provide some new insights into the mechanisms
responsible for high level, tissue-specific expression conferred by the
ß-casein CoRE. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The regulation of ß-casein gene transcription is controlled primarily by a CoRE that integrates signaling from the lactogenic hormones, PRL, insulin, and hydrocortisone, in mammary epithelial cells (reviewed in Ref. 3). The individual transcription factors, which bind to and activate the ß-casein CoRE, have been well characterized in vitro, and in several cases in vivo analysis has been performed as well. Signal transducer and activator of transcription (STAT) 5, glucocorticoid receptor (GR), and CCAATT/enhancer binding protein-ß (C/EBPß) have been identified as important activators of transcription (reviewed in Ref. 4 ; schematic representations of the ß-casein proximal promoter are shown in Refs. 4, 5). The CoRE in the proximal promoter of the ß-casein gene contains a consensus and a nonconsensus binding site for STAT5, at least three binding sites for C/EBP family members, and several half-palindromic binding sites for GR (half- GREs) (6, 7, 8, 9, 10, 11). These half-GREs are closely interspersed with the binding sites for the other transcription factors. Although half-GREs are not the canonical elements known for eliciting GR responses, the importance of these elements in the ß-casein promoter has been unequivocally demonstrated by site-directed mutagenesis (7). Understanding how STATs, GR, and C/EBPs interact with each other and act in a concerted manner should provide a clearer understanding of how the ß-casein CoRE conveys high level, mammary-specific gene expression.
Efforts to understand the mechanism by which transcriptional activity is enhanced by the ß-casein gene CoRE were initiated by the analysis of STAT5 and GR interactions using a COS-7 cell reconstitution system. Direct protein-protein interactions of STAT5 and GR were demonstrated, resulting in transcriptional synergy at the ß-casein promoter (12). Two STAT5 proteins, STAT5a and STAT5b, which are encoded by different genes (13), are both capable of transcriptional synergy with GR (14) and are associated with GR in the mammary epithelium and HC11 mammary epithelial cells (15).
The C/EBPs are a family of transcription factors that contain an amino-terminal transactivation domain, which differs among family members, and a carboxy-terminal basic leucine zipper domain (bZIP) responsible for dimerization and DNA binding, which is more highly conserved among family members (16, 17). Multiple C/EBP isoforms can be generated from the intronless genes, which encode several different C/EBPs by either differential translation start site utilization (18) or selective proteolysis (19). For example, from a single C/EBPß mRNA, at least three transcripts can be translated, two activating isoforms called LAPs (originally identified as liver-enriched activating proteins) and one dominant negative isoform called LIP (originally identified as liver-enriched inhibitory protein) (20). ß-Casein gene expression is reduced 85% to 100% in mammary epithelial cells derived from C/EBP ß knockout (KO) mice (21, 22).
Interaction and transcriptional cooperation between C/EBPß and GR
have been studied in transactivation of several genes, including -1
acid glycoprotein, phosphoenolpyruvate carboxykinase (PEPCK),
and herpes simplex virus thymidine kinase (HSV) (23, 24, 25, 26). Although GR
has been shown to interact with C/EBPß, C/EBP, and C/EBP (24, 25), transcriptional cooperation with GR was demonstrated to be
specific for C/EBPß for at least two of these genes (PEPCK and HSV)
(24, 26).
Transcriptional cooperation between C/EBPß and GR had not been studied previously in ß-casein transactivation. Additionally, transcriptional cooperation between STAT5 and C/EBP family members had not been examined in transactivation of any other gene to our knowledge. Studies were initiated to analyze the potential cooperative effects between STAT5, C/EBPß, and GR on ß-casein gene transcription. Transcriptional cooperation of STAT5, C/EBP family members, and GR on the ß-casein transactivation was shown to be specific to the LAP C/EBPß isoform. Unexpectedly, STAT5 and C/EBPß did not exhibit cooperative effects in the absence of GR. Transcriptional cooperation between the three proteins required full-length GR in a transcriptionally active state. The determinants for STAT5 and GR transcriptional cooperativity were also found to be different from those required for cooperation among all three proteins. These studies have helped elucidate some of the mechanisms involved in transcription factor cooperation in ß-casein transactivation and have provided a better understanding of how the ß-casein CoRE facilitates interactions and increased transcription not observed with the individual transcription factors.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
, lane 2,
and Ref. 13). Under these conditions, cotransfection of STAT5a and GR
followed by treatment with PRL and hydrocortisone (HC) gave a 55% ±
6% SEM increase in ß-casein promoter activity as
compared with STAT5a alone (Fig. 1A, lane 5 and Fig. 1C, lane 8).
Because of the critical role of C/EBPß on ß-casein gene expression
observed in C/EBPß-deficient mammary epithelial cells (21, 22),
studies of the potential cooperative effects between C/EBPß and STAT5
and between C/EBPß and GR at the ß-casein promoter were initiated.
For these studies, an expression construct that expressed only the LAP
isoforms of C/EBPß was used.
|
, lane 4). Surprisingly, cotransfection of STAT5a and
LAP also did not lead to a further stimulation of ß- casein
promoter activity over that found with STAT5a alone (Fig. 1A, lane 6).
When all three transcription factors are cotransfected, however, an
overall increase in the transcriptional activity was observed
reproducibly (Fig. 1A, lane 8). With LAP present, an increase of 38%
± 6% SEM was seen over the level of transactivation
imparted by STAT5a and GR in the absence of LAP, or 107% ± 10%
SEM compared with STAT5 alone. In terms of fold induction,
reporter gene transcription is increased 4.2-fold by STAT5a alone,
6.5-fold by STAT5a + GR, and 8.5-fold by STAT5a + GR + LAP after
hormone treatment over the uninduced level (Fig. 1A, lanes 9–11).
Therefore, the ability of LAP to cooperate with STAT5a was dependent on
the presence of GR. The same response was observed if STAT5b was used
in place of STAT5a (data not shown). Since no difference was seen
between the two highly related STAT5 transcription factors for this
response, all subsequent analyses were performed only with STAT5a.
To determine whether the lack of a cooperative effect between STAT5 and
LAP was due to a limiting amount of LAP expression, titration
experiments were performed. Varying amounts of LAP ranging from 5 ng to
200 ng were cotransfected alone, with STAT5a or with STAT5a + GR. At
all concentrations of LAP tested, GR was necessary for the cooperative
effects with STAT5a (Fig. 1B, lanes 11–14). Cooperativity between
STAT5a and LAP was not seen in the absence of GR at any concentration
of LAP (Fig. 1B, lanes 6–9). Additionally, LAP did not affect the
basal level of transcription from the ß-casein promoter at any
concentration (Fig. 1B, lanes 1–4).
Even before the specific transcription factors responsible for
conveying these effects were identified, it was known that PRL and
glucocorticoids are both essential for ß-casein gene expression (27, 28). Accordingly, the hormonal dependence of the cooperativity between
STAT5, GR, and LAP was examined. In the absence of PRL, no induction of
the ß-casein promoter was seen, regardless of HC treatment (Fig. 1C, odd numbered lanes). Consistent with previous observations,
transcriptional cooperation between STAT5a and GR was dependent upon
both HC and PRL (Fig. 1C, lanes 5–8). The cooperative transcriptional
effects of STAT5a, GR, and LAP were also dependent on both HC and PRL
(Fig. 1C, lanes 13–16). These results confirm that LAP addition to
this reconstitution system mimics the in vivo hormonal
requirements for ß-casein gene transcription.
The Role of the C/EBPß Transactivation Domain
The bZIP domain of C/EBPß is required for interaction of
C/EBPß with GR (25), but the amino-terminal portions of C/EBPß are
crucial for transcriptional cooperation with GR in PEPCK
transactivation (26). In the COS cell reconstitution system, the
transactivation domain of STAT5 is not required for the transcriptional
cooperation between STAT5 and GR (7, 14). Accordingly, the role of the
transactivation domain of C/EBPß in regulating transcriptional
cooperativity in ß-casein transactivation by these factors was
examined. LIP is a naturally occurring, dominant-negative isoform of
C/EBPß (20). When LIP was cotransfected with STAT5a, an inhibition of
transcription was observed (Fig. 2, lane
7). Inhibition by LIP also was observed when LIP was cotransfected with
STAT5a and GR (Fig. 2, lane 10). Therefore, addition of GR does not
circumvent the need for the C/EBPß transactivation domain in
regulating ß-casein gene transcription. Furthermore, LIP inhibited
cooperative transactivation by STAT5, GR, and LAP (Fig. 2, lane 11).
These data are consistent with the observation that LIP markedly
inhibited ß-casein gene expression in CHOk1 cells (our unpublished
results). CHOk1 cells contain endogenous C/EBPß, STAT5, and GR and
are one of the few nonmammary cell lines that can activate milk protein
gene transcription without the addition of exogenous transcription
factors (our unpublished results and Ref. 29).
|
and C/EBP expression constructs. Neither C/EBP nor
C/EBP exhibited cooperative transactivation with STAT5a (Fig. 3A, lanes 2 and 4) or with STAT5a and GR
(Fig. 3A, lanes 6 and 8). The same cooperative effect was seen with
C/EBPß as compared with the C/EBPß/LAP- only expression construct
(Fig. 3A, lane 7). This lack of activity of C/EBP and C/EBP on
the ß-casein promoter reporter construct was particularly interesting
given that both C/EBPs were considerably more active than C/EBPß when
their activity was compared using a multimerized C/EBP binding site
[D9-CAT (chloramphenicol acetyltransferase)] reporter
construct in COS-1 cells (data not shown). This finding of C/EBPß
specificity is consistent with previous reports of the selective roles
of different C/EBPs in mammary gland development (see
Discussion).
|
fused to the b-ZIP domain of C/EBPß
(C/EBPß) or the activation domain of C/EBPß fused to the b-ZIP
domain of C/EBP (C/EBPß). Only C/EBPß elicited a
cooperative effect on ß-casein transactivation with STAT5a and GR
(Fig. 3B, lane 6). Like C/EBP, C/EBPß did not cooperate with
STAT5a and GR (Fig. 3B, lane 5), and neither protein cooperated with
STAT5a alone (Fig. 3B, lanes 1–3). Taken together, these results
indicate that cooperative transcriptional regulation of ß-casein
promoter activity with STAT5 and GR requires, and is specific to, the
N-terminal transactivation domain of C/EBPß.
Regions of GR Needed for Transcriptional Cooperation
The DNA binding domain (DBD) of GR is required for the
protein-protein interaction with C/EBPß (24), and the transactivation
function, TAF-2, in the ligand binding domain of GR is required for
transcriptional cooperation with C/EBPß (24, 25). Transcriptional
cooperation with STAT5 requires the N-terminal portions of GR (14). A
protein-protein interaction domain in GR for STAT5 interaction has not
been mapped, but is thought to reside in TAF-1 (see Fig. 4). Given these observations, experiments
were undertaken to determine whether both portions of the GR molecule
would, therefore, be required for cooperative activation of the
ß-casein reporter construct. Expression constructs for N- and
C-terminal truncations of GR (31) were expressed at comparable levels
to the full-length GR in COS-1 cells (Fig. 4, lane 3 for GR 407–795).
The immunoblot shown in Fig. 4 was probed with an anti-GR antibody
recognizing a C-terminal epitope. This antibody (Fig. 4, lane 2),
therefore, did not detect GR 3–556. An antibody to an epitope in the
DBD of GR was used to verify GR 3–556 production (data not shown).
|
, lanes 5 and
7). Because GR 3–556 is constitutively active, it gave a higher level
of transactivation than wild-type GR when cotransfected with STAT5a
(Fig. 5A, lane 2 vs. lane 5); however, further
transactivation was not seen upon addition of LAP (Fig. 5A, lane 6).
Addition of LAP also did not influence the lack of transcriptional
cooperation between STAT5a and GR 407–795 (Fig. 5A, lane 8). These
data indicate that the regions of GR that are necessary for
transcriptional cooperation between STAT5 and GR and between C/EBPß
and GR in pairs are also necessary for transcriptional cooperation
between all three transcription factors on the ß-casein promoter.
|
, lane 4), was used. This point mutation
diminished the level of transcriptional cooperativity between STAT5a
and GR (Fig. 5B, lane 3; in agreement with Refs. 7, 14). Abolishing
DNA binding of GR did not, however, effect transcriptional cooperation
between STAT5a, GR, and LAP (Fig. 5B, lane 5). Although the
independence of GR transcriptional effects from the DNA binding
activity of GR may be a result of the COS cell overexpression system
(W. Doppler, personal communication), the fact that the transcriptional
cooperativity still occurs emphasizes the importance of the protein-
protein interactions of GR with STAT5 and C/EBPß in regulating
ß-casein gene expression.
The Role of Transactivation by GR
The data thus far suggested several possible mechanisms for the
observed GR- dependent transcriptional cooperation. One possibility was
that GR was playing a structural role. For example, GR might act as a
bridging molecule between STAT5 and C/EBPß, thereby helping to
provide a favorable conformation for cooperative transactivation.
Alternatively, GR could play a transactivational role dependent on the
presence of C/EBPß. To differentiate between these possibilities, the
GR antagonist RU486 and transactivation-deficient GR mutants were
used.
RU486 allows DNA binding of GR but blocks transactivation by
keeping the C-terminal TAF-2 domain of GR in a conformation unable to
interact with the rest of the transcriptional machinery through
coactivators (32). Surprisingly, when STAT5a and GR were cotransfected
and RU486 treatment was performed, the same level of transactivation
was seen as with HC treatment (Fig. 6, lanes 3 and 4). However, when STAT5a and GR were cotransfected with LAP
and treated with RU486, no additional transactivation occurred (Fig. 6, lane 6). GR was cotransfected with a mouse mammary tumor virus
(MMTV)-reporter gene to verify that RU486 did not activate consensus
GREs in this cell system. As expected, RU486 did not activate the
MMTV-reporter construct and inhibited HC activation as well (Fig. 6, gray bars, lanes 7–10). These data suggest that
transactivation by GR is an important aspect of the transcriptional
cooperativity between STAT5, GR, and C/EBPß.
|
. GR 108–317
contains a deletion that eliminates the core of
the TAF-1 domain. It is not transcriptionally active and cannot repress
AP-1-dependent transcription. GR30IIB contains three point mutations,
which were identified using a large-scale mutagenesis and screening
strategy in yeast. Mutation of these three amino acids severely
compromises transactivation by GR but does not affect GR repression of
AP-1-dependent transcription (34). The fact that the repressive ability
of GR is maintained suggests that the overall structure of the protein
is not severely altered by the point mutations. Both proteins were
expressed at levels comparable to the wild-type GR in COS-1 cells (Fig. 4, lanes 6 and 7). In the absence of LAP, GR30IIB was able to cooperate
transcriptionally with STAT5a similar to wild-type GR, while GR
108–317 could not (Fig. 7, lanes 4
and 6). Neither GR mutant was capable of transcriptional cooperativity
with STAT5a and LAP (Fig. 7, lanes 5 and 7). Thus, the use of the GR
antagonist RU486 and GR mutants containing TAF-1 mutations revealed
that the transcriptional activity of GR is not necessary for
cooperative transactivation between STAT5 and GR. However,
transactivation by GR is an essential component of the transcriptional
cooperativity between STAT5, GR, and C/EBPß at the ß-casein
promoter.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
or
C/EBP. The results indicate that the domains of GR necessary for
interaction and transcriptional cooperativity with STAT5 and C/EBPß
individually were also required for cooperativity between all three
proteins. Finally, different determinants are required for
cooperativity between STAT5 and GR alone as compared with STAT5, GR,
and C/EBPß together.
The experiments in this study were performed using a COS-1 cell
reconstitution system. The advantage of this system is its versatility.
Endogenous STAT5, GR, and C/EBPß were not detected by Western
blotting before transfection even when 100 µg of total cellular
protein were analyzed. After transfection, the protein levels of each
transcription factor were easily visualized by Western blotting from as
little as 5 µg of total cellular protein. Additionally, no increase
in reporter gene activity was detected after PRL or hydrocortisone
treatment in the absence of exogenously added transcription factors
(demonstrated in Fig. 1A, lane 1, and Fig. 2, lane 1, and in data not
shown). Any effects observed on ß-casein transactivation were
entirely dependent on the transfection of exogenous transcription
factors. This allowed the comparison of combinatorial effects of intact
transcription factors with the analysis of the transcription factors
containing deletions and point mutations, as well as with chimeric
proteins. One limitation of using this reconstitution system is that a
lower level of hormonal induction of ß-casein promoter activity was
observed than has been reported in other cell systems (10, 27, 29). A
second limitation is that the overexpression of these transcription
factors may force interactions not necessarily sufficient at lower
endogenous levels of these same transcription factors. High expression
may help stabilize protein-protein interactions and obviate the need
for weaker protein-DNA interactions (W. Doppler, personal
communication). Although it may not be possible for a reconstitution
system to mimic all aspects of an in vivo process, the
results reported herein correlate well with observations obtained from
studies performed using mammary epithelial cells, mammary gland
extracts, and knockout mice (4).
This finding that C/EBPß transactivation of ß-casein is GR
dependent agrees with previous studies. In a cytotoxic T cell line,
glucocorticoid induction of ß- casein promoter activity was
dependent on the region of the CoRE containing the C/EBP binding sites
(37). Cooperation with STAT5 and GR was specific to C/EBPß. C/EBP
and C/EBP could not substitute for C/EBPß. This is consistent with
results obtained in mouse models and in experiments using mammary
epithelial cell lines. Four C/EBP binding sites were found in the
ß-casein proximal promoter. Mutation of these binding sites severely
decreased ß-casein promoter activity in stably transfected HC11
mammary epithelial cells. C/EBPß was the predominant protein in
extracts from HC11 cells that bound to the C/EBP sites (6). In mice,
deletion of C/EBPß severely decreased ß-casein gene expression,
while deletion of C/EBP exhibited no effect on ß-casein gene
expression (22). Although some binding of C/EBP to the ß-casein
CoRE was detected using extracts from HC11 cells, this interaction was
minor compared with C/EBPß (6). Additionally, C/EBP levels are
highest in the mammary gland during involution, a stage when ß-casein
expression is down-regulated (38, 39). The ß-casein promoter is one
of only a few promoters on which specificity for a C/EBP family member
has been reported. Transcriptional cooperation with GR is specific for
C/EBPß in transactivation of the PEPCK and HSV promoters as well (24, 26). Synergy with the Sp-1 transcription factor on the CYP2D5 promoter
is also specific for C/EBPß (40). These data suggest that one way of
conferring specificity for individual C/EBP family members is obligate
interaction and cooperation with other proteins at the promoter.
LIP, the dominant negative isoform of C/EBPß, inhibited ß-casein transactivation both in the presence and absence of LAP. The mechanisms of the ß-casein gene repression by LIP are particularly interesting because LIP expression in the mammary gland is high during pregnancy and is severely decreased during lactation (8). These data strongly suggest physiological relevance for repression of the ß-casein gene by LIP. Because LIP has a greater DNA binding affinity than LAP (20), it is likely that LIP inhibits ß-casein transactivation by binding to the C/EBP binding sites in the CoRE, preventing LAP from binding. Elucidation of the mechanism of ß-casein repression by LIP in the absence of LAP requires further experimentation. Several possibilities exist. In the CoRE of the rat ß-casein proximal promoter, one of the C/EBP binding sites overlaps with a nonconsensus STAT5 binding site. This site alone does not bind STAT5 with high affinity (6, 8) but may bind STAT5 as a tetramer in cooperation with the consensus STAT5 binding site. Tetramerization of STAT5 on suboptimal DNA binding sites has been demonstrated for many other promoters (41, 42, 43). Therefore, LIP binding to the overlapping C/EBP site may inhibit transactivation by STAT5 by preventing formation of a tetrameric STAT5 complex. Additionally, the ß-casein proximal promoter contains a Yin Yang-1 (YY1) binding site known to be important for repression of ß-casein gene transcription (44, 45). LIP interacts directly with YY1 (46) and YY1 interacts with several histone deacetylases (47, 48). Therefore, LIP may contribute to the recruitment of histone deacetylases and active repression of ß-casein transactivation.
One surprising result of the experiments described herein was the finding that transcriptional cooperation between STAT5 and GR in the absence of C/EBPß did not require transactivation by GR. RU486-bound wild-type GR and GR with point mutations in TAF-1 did not eliminate the increase in transactivation observed when GR was cotransfected with STAT5 as compared with activation by STAT5 alone. Deletion of amino acids (a.a.) 108–317 of GR did, however, eliminate the cooperative effect. Although coimmunoprecipitation experiments are required for confirmation, this further maps the interaction domain between STAT5 and GR within the region previously shown to be essential (a.a. 1–407) (14). These results suggest that cooperative transcription between STAT5 and GR in this system is the result of a structural effect rather than a transactivational effect of GR. Binding of STAT5 and GR to their adjacent elements on this promoter (half-GREs for GR) may strengthen the STAT5 interaction with the ß-casein promoter, allowing it to exert increased transcriptional effects. This may either result in or be a result of prolonged tyrosine phosphorylation of STAT5 (49). GR also may be mediating chromatin remodeling events, such as those that have been shown to take place on the MMTV promoter even if GR is bound to RU486 (50). However, this seems less likely since the reporter genes in these experiments were transiently introduced into the cells rather than stably integrated into the chromatin.
There are several nonexclusive, testable models that may explain the effects of GR on transcriptional cooperativity between STAT5 and C/EBPß. One model predicts that a component of GR-dependent C/EBPß activation is a required interaction of C/EBPß with GR to help relieve an inhibitory conformation of C/EBPß. Experiments using the D9-CAT reporter gene construct, driven by multimerized C/EBP binding sites (data not shown), support this hypothesis. Using this reporter construct, C/EBPß also exhibited very little activity in COS-1 cells in the absence of GR, but exhibited increased activity when GR was present. The amino-terminal transactivation domain of C/EBPß contains two repression domains, which, through an intramolecular interaction, inhibit transactivation and may decrease the DNA binding of C/EBPß (51, 52). This repression can be relieved by phosphorylation of C/EBPß via several kinase-mediated cascades including ras- activated MAPK cascades (51, 52, 53). This inhibitory conformation can also be relieved by interaction with other proteins (proposed in Refs. 51, 52). The relief of C/EBPß repression by protein-protein interactions has been convincingly demonstrated with the myb protein on the mim-1 promoter. C/EBPß and myb interact (54) and exhibit transcriptional synergy on the mim-1 promoter (55, 56). In CV-1 cells, C/EBPß bound to its cis-regulatory element but was inactive on the mim-1 promoter in the absence of myb. In the presence of myb, transcriptional synergy was seen (57).
The N terminus of C/EBPß interacts with the E1A region of p300/CBP (58). It is likely that this interaction requires the open, non repressed conformation of C/EBPß. In CV-1 cells, p300/CBP exerted minimal effects on the mim-1 promoter with C/EBPß alone. When constitutively activated ras was cotransfected, p300/CBP enhanced transcription by C/EBPß on the mim-1 promoter. Cotransfection of myb without ras allowed the same enhancement by p300/CBP to occur, and the transcriptional synergy previously observed by C/EBPß and myb was greatly enhanced (59). Therefore, a common theme emerges that may explain some of the specificity of expression from a CoRE. Interaction of another protein acting on the CoRE with C/EBPß may relieve the inhibitory conformation of C/EBPß and allow recruitment of transcriptional activators such as p300/CBP.
Another model predicts that the transactivation function of GR contributes a second component to the GR dependence of C/EBPß activation of ß-casein. It appears unlikely that changing the C-terminal conformation of GR by binding RU486 and mutating the N-terminal transactivation domain (TAF-1) would both disrupt the interaction between GR and C/EBPß. Nevertheless, both methods of eliminating GR transactivation abolished transcriptional cooperativity with C/EBPß and STAT5. Additionally, using the D9-CAT reporter, the GR30IIB mutant activated C/EBPß to a similar extent as wild-type GR (data not shown). One possibility is that C/EBPß and GR recruit a coactivator complex together that neither can effectively recruit alone. Boruk et al. (24) have proposed another mechanism to explain the transcriptional cooperativity. They theorized that C/EBPß recruits an activation complex to the HSV promoter after which the activity of this activation complex is enhanced by TAF-2 of GR. This enhancement could be independent of GR binding to DNA or interaction with C/EBPß. It should be noted that this mechanism and the coactivator mechanism postulated above are not mutually exclusive.
Because C/EBPß and GR cannot activate ß-casein gene transcription
in the absence of STAT5, cooperativity between a STAT5-recruited
activation complex and an activation complex recruited jointly by
C/EBPß and GR seems likely. It has been demonstrated that multiple
coactivator activities are required for transactivation by retinoic
acid receptor, hepatic nuclear factor-1, and NF-
B (60, 61, 62). It is
very likely, therefore, that the recruitment of multiple coactivators
is necessary for high level transactivation of many genes. CoREs may
accomplish this by using multiple transcription factors as a way to
impart specificity of gene expression. Analysis of the coactivators
recruited by STAT5, GR, and C/EBPß to the ß-casein promoter and
their contribution to transactivation is an important area of future
investigation.
There are still some aspects of regulation of the ß-casein CoRE observed in vivo that cannot be readily explained by these data. In STAT5a-deficient mice, the level of activated STAT5b is severely reduced, yet ß-casein expression is only marginally affected (63). Analysis of ß-casein expression in STAT5a- and STAT5b-deficient mammary gland transplants has not yet been reported, so it is still not known whether a small amount of STAT5 is sufficient for transactivation in vivo. In contrast, transactivation in cell culture systems is highly dependent on STAT5 (9, 13). Decreasing the level of STAT5 in the COS-1 reconstitution system severely decreased ß-casein transactivation even in the presence of GR and C/EBPß (data not shown). In the reconstitution system, transactivation also is observed in the absence of C/EBPß, while in C/EBPß-deficient mammary epithelial cells, ß-casein expression is severely reduced (by 85–100%) (21, 22). One possibility is that C/EBPß may play an additional role in ß-casein transactivation before formation of the hypothesized activation complexes. C/EBPß was recently found to interact with the SWI/SNF complex, an ATP-dependent chromatin remodeling complex (64), and this may be one possible explanation for the discrepancies between the in vivo and cell culture observations. Modification of the reconstitution system to examine the effects of STAT5, C/EBPß, and GR on the ß-casein transactivation with the reporter construct stably integrated into the chromatin may provide further information on how these transcription factors act on the ß-casein CoRE. Additionally, analysis of GR-deficient mammary epithelial cells may help confirm the importance of GR in transactivation from the ß- casein CoRE.
In summary, several unique roles for GR at the CoRE located in the ß-casein proximal promoter have been observed. GR appears to promote the formation of an activated conformation of C/EBPß as well as prolong the activated state of STAT5 (49). Additionally, it is likely that GR, C/EBPß, and STAT5 together recruit an activation complex to the ß-casein CoRE that cannot be efficiently recruited by any of the transcription factors individually. Reconciliation of the differences between the in vivo and cell culture observations and analysis of the composition and assembly of the proposed activation complexes are important avenues for future investigation.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
3, and
pSCTLIP, were kindly provided by Dr. Ueli Schibler and have been
previously described (20). For the comparative experiments, we used
expression vectors for C/EBP, ß, and , which were potentially
capable of generating multiple isoforms if the alternative downstream
translation start sites were employed, because expression vectors
capable of producing only activating isoforms of C/EBP and C/EBP
were not readily available. Expression vectors for C/EBP, C/EBPß,
and C/EBP in the pSV2sport expression vector as well as expression
vectors for chimeric fusion proteins C/EBP ß and C/EBP ß were
kindly provided by Drs. Gerald Elberg and Sophia Tsai (Baylor College
of Medicine) and have been previously described (30). These cDNAs were
subcloned into the pSCT expression vector using restriction enzyme
sites in the multiple cloning sites of both vectors. The original cDNAs
of C/EBP , C/E 66 Pß, and C/EBP were kindly provided to Drs.
Elberg and Tsai from Dr. Steven McKnight (University of Texas
Southwestern Medical Center, Dallas, TX). Dr. Rainer Lanz
(Baylor College of Medicine) kindly provided the pSTC and pSCT vectors.
Wild-type GR in the pSTC vector and the following GR mutant constructs,
pSTC GR 3–556, pSTC GR 407–795 (X-795), and pSTC GR C482S, were
kindly provided by Dr. Rainer Lanz and Dr. Sandro Rusconi (University
of Fribourg, Fribourg, Switzerland) and have been previously described
(69). Wild-type GR in the p6R vector and GR 108–317 and GR30IIB in
the same vector were kindly provided by Drs. Jorge Iniguez-Lluhi and
Keith Yamamoto (University of California San Francisco) and have been
previously described (34). Because the Rous sarcoma virus (RSV)
promoter drove these vectors and RSV is C/EBPß responsive, these
cDNAs were subcloned into the pCR3.1 expression vector
(Invitrogen, Carlsbad, CA), which has a CMV promoter like
the vectors used for all the other transcription factors. All plasmids
were purified using QIAGEN DNA maxi-prep kits (QIAGEN,
Valencia, CA).
Cell Culture, Transient Transfection, and Reporter Gene
Assays
DMEM, trypsin-EDTA, donor horse serum, and glutamine were
purchased from JRH Biosciences (Lenexa, KS). FBS was
purchased from Summit Biotechnologies (Fort Collins, CO). Gentamicin,
insulin, and hydrocortisone were purchased from Sigma (St.
Louis, MO). RU486 was obtained from Roussel/UCLAF (Romainville,
France). Ovine PRL (lot AFP10692C) was kindly provided by the National
Hormone and Pituitary program (Bethesda, MD). COS-1 cells were obtained
from the ATCC (Manassas, VA). COS-1 cells were routinely
passaged in DMEM + 10% FBS in the presence of gentamicin. COS-1 cell
transfections were performed 1 day after passaging the cells into the
35-mm wells of six-well tissue culture plates. Transfection was
performed using Superfect Reagent (QIAGEN). In each well,
50 ng of pCMVßgal, 50 ng of pECEPRL-R-L, and 200 ng of -2,300/+490
ß- casein LUC were transiently cotransfected with different
combinations of transcription factor expression constructs or the
corresponding empty vectors as controls. Usually, 50 ng of each
transcription factor were used; 2 µg total DNA and 10 µl of
Superfect were used per 35-mm well. Transfections were performed
according to the manufacturer’s instructions. During transfection and
thereafter, the cells were maintained in DMEM + 10% charcoal-stripped
horse serum with gentamicin and 5 µg/ml insulin. Twenty-four hours
after transfection, treatment with hydrocortisone (1 µg/ml), RU486
(1 x 10-7 M) and/or ovine PRL
(1 µg/ml) was performed for 24 h as indicated. Luciferase assays
were performed by standard methods on a MLX microtiter plate
luminometer (Dynex Technologies, Chantilly, VA). Luciferin was
purchased from Molecular Probes, Inc. (Eugene, OR) and
used to make substrate containing 1 mM luciferin, 0.1
M Trizma phosphate, 12 mM
MgCl2, and 2.4 mM ATP.
ß-Galactosidase assays were performed by standard protocols (70).
O- Nitrophenyl ß-D-galactopyranoside
was purchased from Sigma. CAT assays were performed using
a CAT enyzme-linked immunosorbent assay (ELISA) kit (Roche Molecular Biochemicals, Indianapolis, IN) according to the
manufacturer’s instructions.
Statistical Analysis
Univariate ANOVA was used to test for equality of mean relative
light units (RLU)/ßGAL values across treatment groups. The
null hypothesis was that all treatments had the same mean values of
RLU/ßGAL. ANOVA runs were performed using data from triplicate
samples in treatments from a single experiment with necessary
Bonferroni corrections to P values based on the number of
multiple tests. ANOVA runs were also performed using data from all
treatments in all experiments combined. For these analyses, fitted
values of marginal means and their SEs were used
in hypothesis tests for equal means, with necessary Bonferroni
corrections to P values based on the number of multiple
tests. Analysis was performed using the SPSS statistical software
package (SPSS Version 10, SPSS, Inc., Chicago, IL).
Antibodies and Western Blot Analysis
SDS-PAGE and Western blot analysis was performed by standard
protocols that have been previously described (49). STAT5a and GR were
separated on 7.5% running gels and the C/EBPs were separated on 12%
running gels. Affinity purified rabbit polyclonal anti- STAT5a antibody
has been previously described (68). The following anti-GR antibodies
were used: rabbit polyclonal anti-GR (P-20) TransCruz antibody
(Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at a
1:2,000 dilution and the monoclonal anti-GR antibody, BuGR2
(Affinity BioReagents, Inc. Golden, CO) at a 1:600
dilution. Rabbit polyclonal antibodies anti-C/EBP (14AA), C/EBPß
(C-19), and C/EBP (C-22) from Santa Cruz Biotechnology, Inc. were used at a 1:1,000 dilution to detect those
proteins.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This work was supported by NIH Grant CA-16303.
Received for publication September 15, 2000. Revision received November 6, 2000. Accepted for publication November 9, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
are expressed in mammary epithelial cells
and bind to multiple sites in the ß-casein gene promoter. J Biol
Chem 270:17962–17969 and C/EBPß isoform expression. Genes Dev 14:1920–1932 regulates
generation of C/EBPß isoforms through activation of specific
proteolytic cleavage. Mol Cell Biol 19:1695–1704, is essential
for ductal morphogenesis, lobuloalveolar proliferation, and functional
differentiation in the mouse mammary gland. Genes Dev 12:1917–19281-acid glycoprotein gene acute phase
responsive element by multiple isoforms of C/EBP and glucocorticoid
receptor. J Biol Chem 268:15681–156881-acid
glycoprotein gene via direct protein-protein interaction. Mol Cell Biol 13:1854–1862
2 promoter activity by
CCAAT/enhancer-binding proteins. J Biol Chem 275:27815–27822 (C/EBP-delta) and C/EBP-ß mRNA
expression in the involuting mouse mammary gland. Breast Cancer Res
Treat 58:57–63[CrossRef][Medline]
to synergize with an
Sp1 protein is specified by the leucine zipper and activation domain.
Mol Cell Biol 17:2038–2047[Abstract]
gene
transcription. J Biol Chem 272:31821–31828B requires multiple coactivators.
Mol Cell Biol 19:6367–6378This article has been cited by other articles:
 |
|
 |
|
  Y. Su, K. Shankar, and R. C. M. Simmen Early Soy Exposure via Maternal Diet Regulates Rat Mammary Epithelial Differentiation by Paracrine Signaling from Stromal Adipocytes J. Nutr., May 1, 2009; 139(5): 945 - 951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Fang, M. A. Rycyzyn, and C. V. Clevenger Role of c-Myb during Prolactin-Induced Signal Transducer and Activator of Transcription 5a Signaling in Breast Cancer Cells Endocrinology, April 1, 2009; 150(4): 1597 - 1606. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Newton and N. S. Holden Separating Transrepression and Transactivation: A Distressing Divorce for the Glucocorticoid Receptor? Mol. Pharmacol., October 1, 2007; 72(4): 799 - 809. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Xu, V. A. Spencer, and M. J. Bissell Extracellular Matrix-regulated Gene Expression Requires Cooperation of SWI/SNF and Transcription Factors J. Biol. Chem., May 18, 2007; 282(20): 14992 - 14999. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Buser, E. K. Gass-Handel, S. L. Wyszomierski, W. Doppler, S. A. Leonhardt, J. Schaack, J. M. Rosen, H. Watkin, S. M. Anderson, and D. P. Edwards Progesterone Receptor Repression of Prolactin/Signal Transducer and Activator of Transcription 5-Mediated Transcription of the {beta}-Casein Gene in Mammary Epithelial Cells Mol. Endocrinol., January 1, 2007; 21(1): 106 - 125. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. M. Vidal, R. Merino, E. Rico-Bautista, L. Fernandez-Perez, D. J. Chia, J. Woelfle, M. Ono, B. Lenhard, G. Norstedt, P. Rotwein, et al. In Vivo Transcript Profiling and Phylogenetic Analysis Identifies Suppressor of Cytokine Signaling 2 as a Direct Signal Transducer and Activator of Transcription 5b Target in Liver Mol. Endocrinol., January 1, 2007; 21(1): 293 - 311. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Vanselow, W. Yang, J. Herrmann, H. Zerbe, H.-J. Schuberth, W. Petzl, W. Tomek, and H.-M. Seyfert DNA-remethylation around a STAT5-binding enhancer in the {alpha}S1-casein promoter is associated with abrupt shutdown of {alpha}S1-casein synthesis during acute mastitis J. Mol. Endocrinol., December 1, 2006; 37(3): 463 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. B. Kabotyanski, M. Huetter, W. Xian, M. Rijnkels, and J. M. Rosen Integration of Prolactin and Glucocorticoid Signaling at the {beta}-Casein Promoter and Enhancer by Ordered Recruitment of Specific Transcription Factors and Chromatin Modifiers Mol. Endocrinol., October 1, 2006; 20(10): 2355 - 2368. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. J. Seymour, L. E. Roberts, M. A. Fini, L. A. Parmley, T. L. Oustitch, and R. M. Wright Stress Activation of Mammary Epithelial Cell Xanthine Oxidoreductase Is Mediated by p38 MAPK and CCAAT/Enhancer-binding Protein-beta J. Biol. Chem., March 31, 2006; 281(13): 8545 - 8558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Dhawan, X. Peng, A. L. M. Sutton, P. N. MacDonald, C. M. Croniger, C. Trautwein, M. Centrella, T. L. McCarthy, and S. Christakos Functional Cooperation between CCAAT/Enhancer-Binding Proteins and the Vitamin D Receptor in Regulation of 25-Hydroxyvitamin D3 24-Hydroxylase Mol. Cell. Biol., January 1, 2005; 25(1): 472 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Zhang, H. Zhang, L. Lee, and M. M. Ip The Liver-Enriched Inhibitory Protein Isoform of CCAAT/Enhancer-Binding Protein {beta}, But Not Nuclear Factor-{kappa}B, Mediates the Transcriptional Inhibition of {beta}-Casein by Tumor Necrosis Factor-{alpha} Endocrinology, June 1, 2004; 145(6): 2833 - 2844. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Litterst, S. Kliem, D. Marilley, and E. Pfitzner NCoA-1/SRC-1 Is an Essential Coactivator of STAT5 That Binds to the FDL Motif in the {alpha}-Helical Region of the STAT5 Transactivation Domain J. Biol. Chem., November 14, 2003; 278(46): 45340 - 45351. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Bromberg-White, E. Sen, S. Alam, J. M. Bodily, and C. Meyers Induction of the Upstream Regulatory Region of Human Papillomavirus Type 31 by Dexamethasone Is Differentiation Dependent J. Virol., October 15, 2003; 77(20): 10975 - 10983. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Paukku, J. Yang, and O. Silvennoinen Tudor and Nuclease-Like Domains Containing Protein p100 Function as Coactivators for Signal Transducer and Activator of Transcription 5 Mol. Endocrinol., September 1, 2003; 17(9): 1805 - 1814. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. B. Kabotyanski and J. M. Rosen Signal Transduction Pathways Regulated by Prolactin and Src Result in Different Conformations of Activated Stat5b J. Biol. Chem., May 2, 2003; 278(19): 17218 - 17227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-U. Wagner, C. A. Boulanger, M. D. Henry, M. Sgagias, L. Hennighausen, and G. H. Smith An adjunct mammary epithelial cell population in parous females: its role in functional adaptation and tissue renewal Development, March 5, 2003; 129(6): 1377 - 1386. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. E. Timsit and D. S. Riddick Stimulation of Hepatic Signal Transducer and Activator of Transcription 5b by GH Is Not Altered by 3-Methylcholanthrene Endocrinology, September 1, 2002; 143(9): 3284 - 3294. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kingsley-Kallesen, S. S. Mukhopadhyay, S. L. Wyszomierski, S. Schanler, G. Schutz, and J. M. Rosen The Mineralocorticoid Receptor May Compensate for the Loss of the Glucocorticoid Receptor at Specific Stages of Mammary Gland Development Mol. Endocrinol., September 1, 2002; 16(9): 2008 - 2018. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Rycyzyn and C. V. Clevenger The intranuclear prolactin/cyclophilin B complex as a transcriptional inducer PNAS, May 14, 2002; 99(10): 6790 - 6795. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Miyoshi, J. M. Shillingford, G. H. Smith, S. L. Grimm, K.-U. Wagner, T. Oka, J. M. Rosen, G. W. Robinson, and L. Hennighausen Signal transducer and activator of transcription (Stat) 5 controls the proliferation and differentiation of mammary alveolar epithelium J. Cell Biol., November 12, 2001; 155(4): 531 - 542. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Mukhopadhyay, S. L. Wyszomierski, R. M. Gronostajski, and J. M. Rosen Differential Interactions of Specific Nuclear Factor I Isoforms with the Glucocorticoid Receptor and STAT5 in the Cooperative Regulation of WAP Gene Transcription Mol. Cell. Biol., October 15, 2001; 21(20): 6859 - 6869. [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 |
