| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Physiology and Biophysics, University of Iowa College of Medicine, Iowa City, Iowa 52242
Address all correspondence and requests for reprints to: Deborah L. Segaloff, Ph.D., Department of Physiology and Biophysics, University of Iowa College of Medicine, Iowa City, Iowa 52242. E-mail: deborah-segaloff{at}uiowa.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The expression of the LHR in rat granulosa cells is induced by the synergistic actions of FSH and estradiol (5, 6, 7, 8, 9, 10). Primary cultures of granulosa cells isolated from hypophysectomized or immature rats pretreated with diethylstilbestrol (DES; a compound with estrogen-like activity) have been a widely used model system for studying the FSH-mediated induction of the LHR (2, 3, 4). Thus, when cultured in the absence of FSH, few or no LHR are detectable. The addition of FSH or compounds that cause an increase in the levels of intracellular cAMP cause a dose-dependent increase in the expression of the LHR. The changes in the numbers of cell surface LHR are closely paralleled by changes in LHR messenger RNA (mRNA) (11). In both cases, there is a lag time of at least 24 h after FSH [or 8-bromo-cAMP (8-Br-cAMP)] administration before increases in LHR or LHR mRNA become detectable (11). Using nuclear run-on assays, it has been shown that the addition of FSH or 8-Br-cAMP causes an increase in the rate of transcription of the LHR gene after a 24-h lag time, and that the continuous presence of FSH or 8-Br-cAMP is required to maintain elevated levels of rLHR gene transcription and rLHR mRNA (11). Thus, the induction of the LHR gene by FSH/cAMP is due at least in part to an increase in the transcription of the LHR gene.
When inspecting the sequence of the 2.1 kb upstream of the coding sequence of the rLHR gene it is apparent that there are no consensus sequences for cAMP-responsive elements, suggesting that other sequences mediate the FSH/cAMP responsiveness of the LHR gene (12). Thus, the present studies were undertaken to determine those elements responsible for basal activity of the rLHR gene as well as those that mediate the FSH/cAMP responsiveness of the gene. Toward this end, the properties of the rLHR gene in undifferentiated vs. differentiated granulosa cells were compared. It should be noted that all numbering of the rLHR gene is relative to the start of translation, as the rLHR gene contains several transcriptional start sites (12, 13). The studies presented show that the proximal 155 bp confer basal activity. The data presented further show that several regions of the rat LHR (rLHR) 5'-flanking region confer cAMP responsiveness, suggesting that multiple elements act in a combinatorial fashion to mediate cAMP induction of the LHR gene.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Primary cultures of granulosa cells
Immature Sprague Dawley female rats (20–21 days of age, from
Harlan Sprague Dawley, Inc., Indianapolis, IN) were primed
sc with SILASTIC brand capsules (18 mm; Dow Corning Corp.,
Midland, MI) containing 25 mg DES (a nonsteroidal compound with
estrogen-like activity) for 5 days. The ovaries were then excised, and
the granulosa cells were released as previously described (11). The
granulosa cells were plated on fibronectin-coated tissue culture plates
and cultured in growth medium [DMEM-Ham’s F-12 (1:1) supplemented
with 10 mM HEPES (pH 7.4), 50 µg/ml gentamicin, 0.1%
BSA, 1 µg/ml ITS, and 18 ng/ml corticosterone] at 37 C in a
5% CO2 incubator.
DNA reporter gene constructs
The promoterless pLLV3-luciferase plasmid was provided by Dr.
Richard Maurer (Oregon Health Science University, Portland, OR).
The rLHR-luciferase plasmid constructs containing various fragments of
the 5'-flanking region of the rLHR gene subcloned into the polylinker
region of pLLV3-luciferase plasmid were obtained from Dr. Mario Ascoli
(University of Iowa, Ames, IA). These fragments correspond to
nucleotides (nt) -2056 to -1, -1375 to -1, -186 to -1, -155 to
-1, -120 to -1, -90 to -1, -70 to -1, and -40 to -1 relative
to the translation initiation codon, respectively. These constructs
are, respectively, termed rLHR-2056-luc, rLHR-1375-luc, rLHR-186-luc,
and so forth (14).
Mutagenesis
Mutants were generated by PCR using the overlap extension method
(15). The specific templates were obtained from the LHR-2056-luc or
LHR-186-luc plasmid digested with appropriate restriction
endonucleases. PCR-amplified DNA fragments containing mutation sites
were purified and recovered by low melting agarose electrophoresis,
then subcloned into linearized LHR-2056-luc plasmid from which
wild-type fragment had been removed. The entire PCR region for each
mutant was verified by sequence analysis (16). For individual
mutations, the core sequence of Sp1 sites GGGCGG was
converted to GGAGAG. The activating protein-2 (AP-2)
site CCCAGGC was converted to GATCCTG, and the
steroidogenesis factor-1 (SF-1) site AGGTCA was converted
to TAAAGT.
Transient transfections
Granulosa cells were plated in 60-mm dishes at 4 x
106 cells/dish for 16 h, then transiently transfected
by the calcium phosphate precipitation method (17) using 24 µg
plasmid/60-mm dish. After 4 h of exposure to DNA precipitation,
the cells were washed three times with Waymouth’s MB 752/1 medium
supplemented with 0.1% BSA, refed with fresh growth medium, and
cultured for 40 h in the absence or presence of 3 mM
8-Br-cAMP.
Luciferase assays
In preliminary experiments it was determined that maximal
cAMP-inducible luciferase activity was observed 40–48 h after
transfection. For the experiments shown the cells were lysed 40 h
after transfection with 250 µl lysis buffer (25 mM
glycylglycine hydrochloride, 15 mM MgSO4, 4
mM EGTA, 1 mM dithiothreitol, and 0.5% Nonidet
P-40) for 15 min at room temperature. The cell debris was removed by
brief centrifugation, and the supernatant was transferred to a clean
tube. The luciferase assay was performed as previously described (18).
Protein concentrations of the lysates were determined using the
Bradford assay (19). In the initial experiments, we corrected all data
for transfection efficiency by cotransfection of the luciferase
plasmids with a ß-galactosidase expression vector as described
previously (12). This was not routinely done, however, because after
many initial experiments it became clear that this correction did not
affect the results obtained. The data for luciferase reporter
activities are, therefore, reported as light units per mg protein.
Gel mobility shift assays
The granulosa cells released from the ovaries of DES-pretreated
rats were plated on 60-mm fibronectin-coated plates at a cell density
of 4 x 106 cells/dish and were incubated with or
without 3 mM 8-Br-cAMP for 40 h. The nuclear extracts
were then prepared as previously described (14). Protein concentrations
were determined by Bradford assay (19). Typically, the yield was
approximately 6–12 µg nuclear protein/dish. All of the fragments
used in gel shift assays were prepared from the rLHR-2056-luc plasmid
by digestion using a variety of restriction endonucleases. The
fragments thus generated are shown in Table 1
. Three Sp1 site probes were prepared by
annealing complementary oligonucleotides spanning -91/-70,
AGGCCGAGGGGCGGGCAGAGGG (Sp1a); -111/-90, GGGCCGGAGGGCGGGAAGGCAG
(Sp1b); and -151/-130, GGGGTGGGGGGCGGAGAGAGGG (Sp1c). The mutated Sp1
site fragments were synthesized according to the wild-type sequence,
except that the core sequence of the Sp1 site (GGGCGG) was
converted to GGAGAG. The dephosphorylated fragments and Sp1
site probes were labeled by T4 polynucleotide kinase and
[
-32P]ATP and then incubated (10 fmol) with nuclear
extracts (2–6 µg) on ice for 1 h in the presence of 10
mM HEPES (pH 7.8), 100 mM KCl, 1 mM
EDTA, 5 mM dithiothreitol, 3 µg poly(dI-dC), and 10%
glycerol in a 20-µl volume. For competition experiments or supershift
assays, unlabeled competitor DNA (100-fold molar excess) or polyclonal
rabbit antibody to Sp1 and/or Sp3 (1 230 g each) were added
simultaneously with the labeled fragments. The resulting DNA-protein
complexes were resolved by electrophoresis on 5% polyacrylamide
gels.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). After
transfection, the cells were incubated for 40 h in the absence or
presence of 8-Br-cAMP. They were then lysed and assayed for luciferase
activity. In preliminary experiments it was determined that the
induction of luciferase activity in cells transfected with the rLHR
reporter gene constructs incubated with FSH was much less than that in
cells incubated with 8-Br-cAMP (data not shown). These observations are
in contrast to the induction of the endogenous rLHR gene in
untransfected granulosa cells, where FSH is a stronger stimulus than
8-Br-cAMP (11). Although we performed several experiments to determine
the reasons underlying the decreased responsiveness of the transfected
cells to FSH as opposed to 8-Br-cAMP, we cannot yet explain this
discrepancy. Nonetheless, because of these results, all experiments
herein used 8-Br-cAMP to cause differentiation of the granulosa
cells.
|
depicts the luciferase activity observed in control and
8-Br-cAMP-treated granulosa cells transfected with the various
5'-deleted rLHR reporter gene constructs. The basal activity remained
relatively low with constructs containing up to 90 bp upstream of the
translational start site. Constructs containing 120, 155, and 186 bp
showed increased basal activity, with the longer constructs containing
1375 or 2056 bp showing decreased basal activity. The changes in basal
activity can more readily be analyzed in Fig. 1B, in which the data are
expressed as a percentage of the maximal basal activity observed. Thus,
there clearly must be one or more elements between -90 and -120 that
exert a positive effect on basal activity, and there may be an
additional positive element(s) between -120 and -155. The decreased
basal activity between -186 and -1375 suggests the presence of one or
more elements in this region that suppress basal activity.
Like the basal activity, the cAMP-induced luciferase activity also
varied depending upon the length of the 5'-rLHR sequence in the
reporter gene (Fig. 1A). The cAMP responsiveness of the cells
expressing the different constructs can more readily be understood by
examining the fold increase in luciferase activity in cells incubated
in the presence vs. the absence of cAMP (Fig. 1C). Thus, it
is readily seen that the fold increases in reporter gene activity upon
8-Br-cAMP treatment increased gradually with increasing lengths of
5'-flanking sequence of the rLHR gene, with the maximal induction
(3.9-fold) occurring in cells transfected with rLHR-2056-luc. As no
increments were found to be statistically significant, there does not
appear to be a particular region that is especially important for cAMP
inducibility. These results suggest that there are multiple elements
within the 2056 bp upstream of the translational start site of the rLHR
gene that participate in the cAMP-mediated induction of the gene.
Interaction of rLHR gene 5'-flanking DNA with nuclear proteins from
control vs. induced rat granulosa cells
The above experiments examining granulosa cells transfected with
5'-deleted rLHR reporter gene constructs suggest that multiple regions
are involved in both the basal and 8-Br-cAMP-induced expression of the
rLHR gene. Gel mobility shift assays were then performed to further
examine the nuclear proteins binding to the various regions of the rLHR
5'-flanking region. Thirteen overlapping probes, which encompass 2.1 kb
of 5'-flanking region of the rLHR gene, were prepared (Table 1). Each
probe was then radiolabeled and incubated with nuclear extracts derived
from control or 8-Br-cAMP-induced granulosa cells, and the DNA-protein
complexes were resolved by PAGE. As shown in Fig. 2, numerous nuclear proteins bound to
each of the different probes, except probe 11, regardless of whether
the proteins were isolated from control or induced granulosa cells. No
detectable complexes were observed associated with probe 11, indicating
the absence (or very low affinity binding) of granulosa cell nuclear
proteins binding to nt -1557 to -1786 of the rLHR gene. In examining
the complexes formed with the other probes, it is apparent that within
a given rLHR region some complexes are the same regardless of whether
the nuclear proteins were isolated from control vs. induced
cells. This is particularly evident with probe 1, where most (or all)
of the complexes detected using extracts prepared from control cells
were also present in extracts prepared from induced cells. These data
suggest that the proteins in the complexes observed in the mobility
shift assay of probe 1 probably mediate the basal regulation of rLHR
gene expression.
|
Because each rLHR 5'-flanking DNA probe overlaps with its neighboring
probes, there may be some proteins that bind to the overlapping regions
and are thus not apparent in experiment shown in Fig. 2. To address
this question, each of the 13 probes was labeled separately and was
competed with the corresponding unlabeled probe or each of the adjacent
unlabeled probes. The resulting gel shift assay indicated that at least
15 proteins were induced after 8-Br-cAMP treatment (data not
shown).
All three Sp1 binding sites are important for both basal and
cAMP-induced expression of the rLHR gene
Because the data reported thus far suggest that multiple elements
throughout the 2.1-kb 5'-flanking region of the rLHR participate in the
cAMP-mediated induction of the rLHR gene, we chose to initially
determine the role, if any, of known consensus elements present in this
region. As illustrated in Fig. 3, the
5'-flanking region of the rLHR gene contains three Sp1 sites located at
nt -83 (Sp1a), -103 (Sp1b), and -143 (Sp1c) relative to the
translational initiation codon, respectively. To determine the
potential roles these Sp1 elements may play in the regulation of rLHR
gene transcription, we individually and combinatorially mutated the
three Sp1 sites in the rLHR-2056-luc construct and examined transfected
granulosa cells for basal and 8-Br-cAMP-inducible reporter gene
activities. As shown in Fig. 4, mutation
of any of the three Sp1 sites resulted in decreased basal reporter gene
activity. Relative to the control wild-type rLHR-2056-luc, constructs
in which any of the three Sp1 sites were mutated exhibited only
45–63% of control activity. Mutations of any two Sp1 sites
simultaneously caused a further decrease in activity to 29–39% of the
control value, and the combined mutations of all three Sp1 sites
decreased basal reporter gene activity further to 17% of the control
value. These results suggest that all three Sp1 sites contribute in a
concerted mechanism to the basal transcription of the rLHR gene.
|
|
, 8-Br-cAMP-induced luciferase activity in cells
transfected with rLHR-2056-luc was markedly reduced by mutations of the
Sp1 sites. The pattern observed for cAMP-inducible expression was
similar to that of basal expression, in that mutation of any one site
caused a decrease in activity, and mutations of two sites or three
sites simultaneously caused even greater reductions. However, in each
case, the magnitude of the reduction in activity was far greater for
the cAMP-inducible expression than for the basal expression. Thus,
mutations of Sp1a, Sp1b, or Sp1c sites decreased reporter gene activity
in response to 8-Br-cAMP to 12–24% of the control value, mutations of
two Sp1 sites simultaneously decreased it to 7–12% of the control
value, and mutations of all three Sp1 sites reduced it to 4% of the
control value. If one examines the data as fold increases in luciferase
activity in cells incubated with vs. without 8-Br-cAMP (Fig. 4C), there was a 5.8-fold increase in activity in cells transfected
with the wild-type rLHR-2056-luc construct. In contrast, cells
transfected with the rLHR-2056-luc construct containing single
mutations of the Sp1 sites responded with only 1.6- to 2.2-fold
increases in activity. The fold induction in luciferase activity was
reduced further only slightly by the simultaneous mutation of either
two or three Sp1 sites. These results suggest that all three Sp1 sites
are involved in the cAMP-mediated induction of the rLHR in rat
granulosa cells, and the integrity of all three sites is required for
induction of the gene by cAMP.
Binding of Sp1 family members to Sp1-binding sites in the rLHR
promoter
Gel mobility shift assays were then performed to determine the
nature of complexes formed between each of the three Sp1 sites and
nuclear proteins from either uninduced or induced granulosa cells. The
data presented in Fig. 5 show the
complexes formed with labeled probes corresponding to Sp1a (Fig. 5A),
Sp1b (Fig. 5B), and Sp1c (Fig. 5C) binding sites. Examination of lanes
1 and 8 of each panel show that with Sp1a, Sp1b, or Sp1c probe, two
complexes are formed. We have designated the complex at the upper
portion of the gel complex I and the one below it complex II. It should
be noted that relative to the gel shifts using the Sp1a and Sp1c
probes, the gel shifts using the Sp1b probe required greater amounts of
nuclear protein and longer exposure times to observe any complex
formation. Therefore, the Sp1b site may be of a lower affinity than the
Sp1a and Sp1c sites. Interestingly, however, when comparing the
complexes formed with each probe, it can be seen that the migration and
intensity of the complexes are the same regardless of whether nuclear
extracts from uninduced cells or 8-Br-cAMP-induced granulosa cells are
used.
|
, lanes 2 and 9) or an oligonucleotide corresponding to
a mutated Sp1-binding site sequence (Fig. 5, lanes 3 and 10) were used.
For probes corresponding to Sp1a, Sp1b, or Sp1c, both complexes I and
II were competed by the wild-type Sp1 oligonucleotide, but not with the
oligonucleotide containing a mutation of the core Sp1-binding sequence.
Both complexes can also be eliminated by Sp1 consensus DNA (data not
shown), suggesting the involvement of an Sp1 family member(s) in the
formation of complexes I and II.
Supershift assays were further used to determine the identities of
complexes I and II. When incubated with labeled probes corresponding to
the Sp1a, Sp1b, or Sp1c sites in the presence of control or induced
nuclear extracts of granulosa cells, antibody to Sp1 decreased the
formation of complex I, while also forming a faint supershifted band
just above complex I (Fig. 5, lanes 5 and 12). There was no reduction
in the intensity of either complex I or II when normal rabbit IgG was
used instead (Fig. 5, lanes 4 and 11), confirming the specificity of
the interaction of the Sp1 antibody with complex I. These data,
therefore, suggest that complex I contains Sp1. Antibody to Sp3
eliminated the formation of complex II (Fig. 5, lanes 6 and 13),
suggesting that complex II contains Sp3. Therefore, both Sp1 and Sp3
bind to each Sp1-binding site in the rLHR gene. To our surprise, the
addition of antibodies to Sp1 and Sp3 together almost totally abolished
the formation of both complexes I and II (Fig. 5, lanes 7 and 14).
However, in examining complex I more carefully, it can be seen that the
Sp1 antibody inhibited the formation of the upper portion of complex 1
(Fig. 5, lanes 5 and 12), whereas the Sp3 antibody inhibited the
formation of the lower portion of the complex I (Fig. 5, lanes 6 and
13). These results suggest that complex I is actually a composite of
two complexes, one formed by the binding of Sp1 and one formed by the
binding of Sp3.
Inhibitory functions of SF-1- and AP-2-binding sites on basal
transcription of the rLHR gene
The 5'-flanking region of the rLHR gene also contains consensus
sequences corresponding to binding sites for SF-1 at nt -174 and -835
and for AP-2 at -835 (Fig. 3). To examine the potential roles of these
factors in regulating basal and cAMP-induced rLHR gene expression, the
following experiments were performed. Using the rLHR-2056-luc reporter
gene construct as a template, each of the two SF-1 sites was
individually mutated. Luciferase assays of granulosa cells transfected
with these constructs showed that mutation of either the SF-1a site (nt
-174) or the SF-1b site (nt -835) had no effect on the promoter
activity in the induced granulosa cells (Fig. 6). However, mutation of the SF-1a site
increased basal promoter activity by 50% (Fig. 6). These results
suggest that SF-1 could play a role in repressing rLHR gene expression
in the basal state, but it does not appear to be involved in the cAMP
induction of the gene.
|
show that mutation of the AP-2 site increased basal reporter gene
activity by 50% compared with that of the wild-type construct. The
fold induction by 8-Br-cAMP was slightly reduced (from 2.0- to
1.7-fold), although the absolute activity of the AP-2 mutant in induced
cells was increased somewhat (Fig. 7). Taken together, these results
clearly demonstrate that although SF-1 and AP-2 may repress basal rLHR
gene expression, they are not required for cAMP-induced expression of
the rLHR gene.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present studies were undertaken to delineate the promoter and regulatory regions involved in basal and cAMP-induced transcription of the rLHR gene in granulosa cells from DES-pretreated rats. In the present studies we focused on the region 2.1 kb upstream of the translation initiation codon. Interestingly, in preliminary experiments it was observed that granulosa cells transfected with a luciferase reporter construct containing this portion of the 5'-flanking region of the rLHR were far more responsive to 8-Br-cAMP than to FSH. This is in contrast to our earlier studies in which, using nuclear run-on assays, it was observed that FSH caused a greater fold increase in the transcription of the endogenous rLHR gene than 8-Br-cAMP (11). When mock-transfected cells or cells transfected with the empty luciferase vector were assayed for [125I]hCG binding after incubations with FSH or 8-Br-cAMP, it was found that both could induce rLHR expression in the transfected cells, with FSH inducing higher levels of receptor than 8-Br-cAMP (data not shown). Therefore, the decreased activity of FSH compared with that of 8-Br-cAMP in stimulating a reporter gene construct driven by nt -2056 to -1 of the rLHR gene suggests the existence of FSH-responsive elements (that may or may not be cAMP dependent) in other regions of the gene (i.e. further upstream, intronically, or in the 3'-flanking region). Further studies are required to address this question.
We have shown herein that the region between nt -155 and -1 represents a minimal promoter for full basal activity of the reporter gene. By mapping 5'-deleted rLHR-luciferase fusion genes, a positive regulatory domain(s) was identified between nt -90 and -120. Another may be located between nt -120 and -155. A negative regulatory domain(s) was identified between nt -186 and -1375. The ability of 8-Br-cAMP to cause increased reporter gene activity in the cells transfected with these constructs suggests that multiple domains within nt -1 to -2056 of the 5'-flanking region of the rLHR gene participate in the cAMP-induced regulation of this gene. Gel mobility shift assays using probes corresponding to overlapping fragments of the 5'-flanking region of the rLHR gene further show numerous cAMP-inducible proteins binding throughout the region. These data suggest that the activation of rLHR gene transcription in rat granulosa cells involves the combinatorial effects of proteins binding to multiple sites within the 5'-flanking region of the gene.
The generally accepted model for the transcriptional regulation of TATA-containing promoters is that the TATA-binding protein (TBP) binds to the TATA box, recruits the general transcription factors, and then positions RNA polymerase to initiate transcription. It has further been shown that the TBP is required for the initiation of transcription of the TATA-less promoter (24, 25). Puph and Tjian have examined the requirement for Sp1 activation of transcription of synthetic TATA-containing and TATA-less promoters and found that Sp1 acts to recruit TFIID to the TATA-less promoter (25). The same set of basic initiation factors is required in the presence and absence of a TATA sequence, including TBP-associated factor (24, 26, 27, 28, 29). The rLHR gene promoter has all the features of TATA-less promoters (30). Thus, it lacks a TATA box, has a GC-rich region with a cluster of Sp1 sites, and exhibits multiple transcriptional start sites (nt -14, -19, and -33) (12, 13).
The present studies using 5'-deletion reporter gene constructs and
reporter genes containing Sp1 mutations suggest that all three Sp1
sites contribute in a concerted mechanism to the basal transcription of
the rLHR gene.2 As shown in
Fig. 4A, mutation of the Sp1a site within the rLHR-2056-luc construct
decreased basal transcription activity by 55%, indicating that the
Sp1a site is essential in the formation of a transcription initiation
complex. However, the Sp1a site alone is not sufficient to initiate
transcription, as the rLHR-90-luc construct, which contains the Sp1a
site but lacks both the Sp1b and Sp1c sites, exhibited low levels of
basal gene reporter activity (5% of the maximal basal activity
observed by rLHR-155-luc and 24% of the basal activity of
rLHR-2056-luc). The construct rLHR-120-luc, which contains the Sp1a and
Sp1b sites, showed moderate basal activity and the subsequent addition
of the Sp1c site, which is included in the rLHR-155-luc construct,
exhibited maximal basal activity. Taken together, the data shown in
Figs. 1B and 4A suggest that the synergistic actions of the three Sp1
sites are responsible for maximal basal transcription of the rLHR gene.
The data further suggest that the Sp1a site at nt -83 acts perhaps by
recruiting the TFIID complex to initiate rLHR gene transcription in
granulosa cells, and the Sp1b and Sp1c behave as activators that
further enhance basal transcription. The supporting role for the Sp1b
and Sp1c sites is further born out by the results showing that mutation
of either site in the context of the rLHR-2056-luc construct reduced
basal activity, but not to as great an extent as mutation of the Sp1a
site. Mutations of any two Sp1 sites decreased basal activity much
further, and mutation of all three Sp1 sites decreased basal activity
to 17% of the control value, suggesting a concerted role in the
actions of these three sites in regulating basal rLHR
transcription.
We assume that the FSH (cAMP)-induced regulation of the rLHR gene is based on this basic machinery as well, as mutation of any of the three Sp1 sites greatly reduced reporter gene activity in 8-Br-cAMP-treated granulosa cells. Although other sites are probably also involved in the cAMP-mediated induction of the rLHR gene in granulosa cells, clearly each of the Sp1 sites plays a major role.
Sp1 is widely expressed and was initially thought to be involved in the transcriptional activation of genes in a nonhormonally regulated manner. Sp3 is a Sp1 family member that has been shown to antagonize Sp1 activity (31) and also to activate transcription (32). In recent years it has been shown that Sp1 and/or Sp3 are indeed involved in the hormone-regulated induction of a variety of genes. For example, the genes for cholesterol side-chain cleavage cytochrome P450 (33, 34), rhesus GH variant (35), serum/glucocorticoid-inducible kinase (36) and LH ß-subunit (37) have been shown to require Sp1-binding sites for cAMP-mediated induction. The results of our experiments extend these observations and provide the first evidence that the Sp1 family and its binding sites are involved in the cAMP-mediated induction of rLHR gene transcription in rat granulosa cells. The mechanism by which the binding of Sp1 family members to the rLHR gene promoter activates transcription in response to 8-Br-cAMP is unclear. As has been shown for other genes (34, 35, 36, 37), we show herein that there is no significant alteration in the DNA-binding activity of Sp1 and/or Sp3 to the rLHR gene promoter upon 8-Br-cAMP stimulation. It is possible that the induction of granulosa cells with 8-Br-cAMP may lead to posttranslational modification of Sp1 and Sp3 that results in changes in their interactions with other transcription factors binding to the rLHR gene promoter. For example, modifications of Sp1 and Sp3 by 8-Br-cAMP treatment may result in the release of a repressor(s) and/or the recruitment of a coactivator(s) that enhance the transcription of the rLHR gene in the granulosa cells. Interestingly, the binding of Sp3 in nuclear extracts from control or induced granulosa cells contributes to the formation of two complexes with different mobilities. One possible explanation for these results is that Sp3 variants with differing mol wt may be binding to the Sp1 sites (38). Another possibility is that the slower migrating complex II may contain an additional protein that interacts with Sp3. The observation that there was no supershifting with the Sp3 antibody and only slight supershifting with the Sp1 antibody further suggests that the interaction of the Sp1 and Sp3 proteins with their respective antibodies inhibits the binding of the proteins to the Sp1 sites.
Although the binding of nuclear Sp1 and Sp3 proteins to the three Sp1-binding sites in the proximal region of the rLHR gene is unaffected by pretreatment of the cells with 8-Br-cAMP, the binding of other nuclear proteins to the 5'-flanking region of the rLHR gene is indeed enhanced by 8-Br-cAMP pretreatment. Taken altogether, these observations suggest that there are other transcription factors binding to different sites (whose binding is affected by cAMP) that may be acting in a combinatorial fashion with Sp1 and Sp3 to cause cAMP induction of the rLHR gene.
The involvement of AP-2- and SF-1-binding sites in activating the transcription of certain genes has been well demonstrated (39, 40, 41, 42, 43, 44). SF-1 has also been shown to act as a key determinant of endocrine development and function (45, 46, 47). Our results, however, suggest that both AP-2- and SF-1-binding sites negatively regulate rLHR gene transcription in granulosa cells. A similar inhibitory effect of AP-2 on basal rLHR gene expression has been previously observed in cell lines derived from a mouse Leydig cell tumor (14, 48). This is the first report to our knowledge of the inhibition of basal rLHR gene transcription by SF-1-binding sites. The mechanisms by which SF-1 positively regulates certain granulosa cell genes, but negatively regulates others (i.e. the rLHR gene) remain to be determined.
The rLHR is expressed in females primarily in ovarian granulosa and luteal cells and in males in testicular Leydig cells. Unlike granulosa cells, which do not express cell surface LHR in the undifferentiated state and require FSH/cAMP and estradiol for expression in the differentiated state, Leydig cells constitutively express LHRs. Furthermore, the addition of cAMP to Leydig cells causes the down-regulation of the LHR, which has been shown to be due to a cAMP-mediated decrease in LHR gene transcription (49). Despite the differences in the actions of cAMP on LHR gene transcription in Leydig cells vs. granulosa cells, it may be instructive to compare the two cell types. The transcriptional regulation of the rLHR gene has been reported by Ascoli and co-workers using MA-10 cells (a clonal line of mouse Leydig tumor cells that expresses the LHR) (14) and by Dufau and co-workers using mLTC cells (a different clonal line of Leydig tumor cells isolated from the same tumor as the MA-10 cells) (13, 48, 50). In studies by both groups of investigators, it was found that the proximal 173 bp of the 5'-flanking region of the rLHR contained a strong promoter. These results are in general agreement with our observations from uninduced granulosa cells that this same region of the rLHR gene provides the elements necessary for maximal basal rLHR transcriptional activity in rat granulosa cells. Deoxyribonuclease protection assays in MA-10 cells demonstrated the binding of Leydig cell nuclear proteins to each of the three Sp1 sites in the proximal promoter; however, the functional consequences of the binding to each site were not addressed in this study. The functional roles of the three Sp1 sites have been investigated by Dufau and colleagues using mLTC cells. Although the results of the Sp1a site at nt -83 are similar to those we observed in rat granulosa cells, the functional roles and binding properties of the Sp1b and Sp1c sites at nt -103 and -143, respectively, are different in the two cell types. Our results show that the Sp1b site plays an important role in activating the basal transcription of the rLHR gene in rat granulosa cells. However, this site does not appear to affect the promoter activity of the rLHR gene in the mLTC mouse Leydig cells (48). The Sp1c site was found to be important for basal expression of the rLHR gene in the mLTC cells. However, it was shown that Sp1 bound to the upstream sequence GGGGTGGGG (from nt -151 to -143) instead of the canonical GC box (GGGCGG, from nt -143 to -128), and that mutation of the canonical GC box did not affect promoter activity. Instead, mutation of the upstream sequence (from nt -151 to -143) caused a reduction in basal promoter activity (50). Our studies of rLHR gene expression in rat granulosa cells, however, indicate that Sp1 binds to the canonical GC box of the Sp1c site, as the complex formed between nuclear extracts and a Sp1c probe can be competed by an oligonucleotide containing the wild-type Sp1-binding sequence, but not by an oligonucleotide in which the canonical GC box has been mutated. Functional assays in the granulosa cells further showed that mutation of this GC box in the Sp1c site caused a reduction in basal promoter activity. In additional experiments, we found that the adjacent sequence upstream of the Sp1c site (GGGGTGGGG) has little effect on the basal promoter activity of the rLHR gene in rat granulosa cells, but it is important in the 8-Br-cAMP-induced expression of the gene in these cells (Chen, S., and D. L. Segaloff, manuscript in preparation).
In summary, our results show that multiple cis elements are involved in both basal and cAMP-mediated transcriptional regulation of the rLHR gene in rat granulosa cells. Furthermore, the three Sp1 sites in the proximal portion of the 5'-flanking region of the rLHR gene appear to be critical for both basal and cAMP-induced transcription of the gene. Additional studies are underway to further identify the other cAMP-responsive elements that are involved in the FSH/cAMP-induced regulation of rLHR gene transcription in granulosa cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 It is important to note that although a
qualitative comparison of basal activities shown in Figs. 1 and 4 is
instructive, a quantitative comparison is difficult because the data in
Fig. 1 are expressed as the percent maximal basal expression (where
maximal activity is observed with rLHR-155-luc) and the data in Fig. 5
are expressed as the percentage of control basal activity observed with
the rLHR-2056-luc template.
Received October 20, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase/c17–20 lyase). Mol Endocrinol 10:147–158This article has been cited by other articles:
 |
|
 |
|
  A.-M. O'Carroll, S. J Lolait, and G. M Howell Transcriptional regulation of the rat apelin receptor gene: promoter cloning and identification of an Sp1 site necessary for promoter activity J. Mol. Endocrinol., February 1, 2006; 36(1): 221 - 235. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Youn, Y. Koo, I. Ji, and T. H. Ji An Upstream Initiator-Like Element Suppresses Transcription of the Rat Luteinizing Hormone Receptor Gene Mol. Endocrinol., May 1, 2005; 19(5): 1318 - 1328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y Kamei, Y Aoyama, T Fujimoto, N Kenmotsu, C Kishi, M Koushi, S Sugano, K Morohashi, R Kamiyama, and R Asakai A steroidogenic cell line with differentiation potential from mouse granulosa cells, transfected with Ad4BP and SV40 large T antigen genes J. Endocrinol., April 1, 2005; 185(1): 187 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sekiguchi, T. Mizutani, K. Yamada, T. Yazawa, H. Kawata, M. Yoshino, T. Kajitani, T. Kameda, T. Minegishi, and K. Miyamoto Transcriptional Regulation of the Epiregulin Gene in the Rat Ovary Endocrinology, December 1, 2002; 143(12): 4718 - 4729. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hamalainen, J. Kero, M. Poutanen, and I. Huhtaniemi Transgenic Mice Harboring Murine Luteinizing Hormone Receptor Promoter/{beta}-Galactosidase Fusion Genes: Different Structural and Hormonal Requirements of Expression in the Testis, Ovary, and Adrenal Gland Endocrinology, October 1, 2002; 143(10): 4096 - 4103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Yoshino, T. Mizutani, K. Yamada, M. Tsuchiya, T. Minegishi, T. Yazawa, H. Kawata, T. Sekiguchi, T. Kajitani, and K. Miyamoto Early Growth Response Gene-1 Regulates the Expression of the Rat Luteinizing Hormone Receptor Gene Biol Reprod, June 1, 2002; 66(6): 1813 - 1819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ascoli, F. Fanelli, and D. L. Segaloff The Lutropin/Choriogonadotropin Receptor, A 2002 Perspective Endocr. Rev., April 1, 2002; 23(2): 141 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Zhang and M. L. Dufau EAR2 and EAR3/COUP-TFI Regulate Transcription of the Rat LH Receptor Mol. Endocrinol., November 1, 2001; 15(11): 1891 - 1905. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Hamalainen, M. Poutanen, and I. Huhtaniemi Promoter Function of Different Lengths of the Murine Luteinizing Hormone Receptor Gene 5'-Flanking Region in Transfected Gonadal Cells and in Transgenic Mice Endocrinology, June 1, 2001; 142(6): 2427 - 2434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chen, X. Liu, and D. L. Segaloff Identification of an SAS (Sp1c Adjacent Site)-Like Element in the Distal 5'-Flanking Region of the Rat Lutropin Receptor Gene Essential for Cyclic Adenosine 3',5'-Monophosphate Responsiveness Endocrinology, May 1, 2001; 142(5): 2013 - 2021. [Abstract] [Full Text]
|
|
|
|
|
|
S. Chen, X. Liu, and D. L. Segaloff A Novel Cyclic Adenosine 3',5'-Monophosphate-Responsive Element Involved In the Transcriptional Regulation of the Lutropin Receptor Gene in Granulosa Cells Mol. Endocrinol., September 1, 2000; 14(9): 1498 - 1508. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
0.05) compared with the induction observed in
control cells transfected with the empty vector. The data shown in B
and C are the mean ± SEM of at least three
independent experiments using different preparations of plasmids.
