| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinologie Cellulaire et Moléculaire de la Reproduction Université Pierre et Marie Curie Centre National de la Recherche Scientifique ESA 7080, Paris, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
T3–1
and LßT2 cell lines, the existence of a distal enhancer [GnRH-R-
specific enhancer (GnSE)] that is highly active in the context of
the GnRH-R gene promoter. We show that the GnSE activity
(–1,135/–753) is mediated through a functional interaction with a
proximal region (–275/–226) that includes the SF-1 response element.
Regions of similar length containing either the AP-1 or GRAS elements
are less active or inactive. Transfection assays using an artificial
promoter containing two SF-1 elements fused to a minimal PRL promoter
indicate that SF-1 is crucial in this interaction. In addition, by
altering the promoter with deletion and block- replacement mutations,
we have identified the active elements of GnSE within two distinct
sequences at positions –983/–962 and –871/–862. Sequence
analysis and electrophoretic mobility shift experiments suggest that
GnSE response elements interact, in these two regions, with GATA- and
LIM-related factors, respectively. Altogether, these data establish the
importance of the GnSE in the GnRH-R gene expression and reveal a novel
role for SF-1 as a mediator of enhancer activity, a mechanism that
might regulate other SF-1 target genes. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit and the specific ß-subunit
of LH and FSH. In addition to this action at the gene level, GnRH also
stimulates the release of LH and FSH, which, in turn, through the
systemic circulation, orchestrate gonadal function including steroid
hormone synthesis and release, as well as gametogenesis (reviewed in
Refs. 1, 2, 3, 4). Both the specificity and magnitude of the pituitary response to GnRH are highly dependent on GnRH receptor (GnRH-R) gene expression. Although the mature anterior pituitary is composed of at least five different endocrine cell types, only gonadotropes express the GnRH-R, which confers cell-specific regulation of gonadotropin secretion. Furthermore, the sensitivity of the pituitary gland to hypothalamic GnRH inputs is dependent on the number of active cell-surface GnRH-Rs, which is itself regulated, at least in part, at the transcriptional level. Thus, molecular mechanisms underlying gonadotrope-specific expression and transcription efficacy of the GnRH-R gene are critical for the normal functioning of the pituitary- gonadal axis.
In addition, the GnRH-R gene is expressed, although to a lower extent, in other tissues such as the hippocampus and the hypothalamus (3) as well as in multiple rat ovarian compartments, especially the granulosa cells of atretic follicles (5, 6, 7). GnRH-R mRNAs have also been detected in Leydig cells, and mature and fetal testes and ovaries as well as in human breast and placental trophoblasts during pregnancy (8; Ref. 9 and references therein; 10–13). More recently, the presence of mRNA encoding GnRH-R has also been described in human peripheral blood mononuclear cells (14). Such findings raise questions about the nature of cis-regulatory elements and cognate trans-acting factors that confer either gonadotrope or extrapituitary expression to the GnRH-R gene.
To investigate this issue, the 5'-flanking sequences of the ovine, human, mouse, and rat GnRH-R gene have been isolated and partially characterized (15, 16, 17, 18, 19, 20, 21). The human, mouse, and rat promoters display strong sequence homology in a region extending over 1,200 bp upstream of the ATG codon. However, the rat and mouse promoters diverge extensively from the human promoter in the position of the transcription start sites. While transcription start sites have been identified within a region that extends over 110 bp upstream of the ATG codon of the mouse and rat promoters, the start sites of the human promoter are clustered at 0.7 and 1.4 kb upstream of the ATG codon.
The gonadotrope-derived T3–1 cell line expresses the GnRH-R
and the -subunit of gonadotropin hormones and has been extensively
used for testing the cell-specific expression of the GnRH-R gene by
transient transfection assays. Analysis of the mouse promoter has led
to the identification of cis-acting elements localized
essentially within 500 bp upstream of the ATG codon. These elements
include a new element termed the GnRH-R-activating sequence or GRAS, a
consensus activator protein-1 element (AP-1), and the
gonadotrope-specific element or GSE (22) that binds the nuclear orphan
receptor, steroidogenic factor-1 (SF-1) (23, 24). This tripartite basal
enhancer appears to be sufficient in vitro to ensure maximum
gonadotrope-specific activity of the mouse promoter. The GRAS element
has also been shown to mediate autocrine/paracrine stimulation of
cell-specific expression of the GnRH-R gene by activin (25). In
parallel, the AP-1 response element has been recently demonstrated to
play a central role in conferring GnRH responsiveness in the murine
GnRH-R gene (26, 27). The gonadotrope-specific activity of the human
promoter is also dependent on an SF-1 response element located in the
5'-untranslated region downstream of the transcription start sites
(28).
Elements analogous to the mouse GRAS, AP-1, and SF-1 are present at corresponding positions in the rat GnRH-R promoter gene. Nevertheless, despite the presence of these cis-acting elements, the promoter of the rat GnRH-R gene seems to be regulated in a different manner. For maximal gonadotrope-specific activity the presence of additional distal elements localized within the –1,150 to –750 bp region (21) are necessary. The distal elements appear to be active only in the context of the GnRH-R promoter. In fact, we have previously reported that these distal elements do not display any activity if they are directly fused to the heterologous minimal thymidine kinase (TK) promoter. However, when the full-length 1.2-kb promoter was fused to the TK promoter, cell-specific activity was recovered. We thus hypothesized that the activity of the upstream elements, thereafter referred to as the GnRH-R-specific enhancer (GnSE), necessitates the presence of promoter-specific elements located in the proximal part of the rat promoter.
In the present study, by using deletion and mutational analysis
combined with functional transfection studies in the murine
gonadotrope-derived T3–1 and LßT2 cell lines (29, 30, 31), we show
that the activity of the GnSE (–1,135/–750) requires sequences
localized within the proximal region (–412 to –26). A critical
element lies at position –245/–237 and contains the consensus
sequence for the SF-1 element. The functional interaction between the
GnSE and proximal elements is confirmed by the demonstration that while
the GnSE and a proximal 50-bp region including the SF-1 element are
capable of independently conferring a more or less modest activity to a
heterologous minimal PRL promoter, both elements induced full
synergistic stimulation. The regulatory elements within the GnSE have
been restricted to two short sequences of 10 and 20 bp extending from
–871 to –862 and from –983 to –962, respectively.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
).
This promoter, thus deleted from an intermediate domain extending from
–753 to –412, was inserted into the pCAT basic vector before the
chloramphenicol acetyltransferase (CAT) gene. In addition, we designed
two similar CAT constructs that contained either the 5' region
(–1,135/–900) or the 3' region (–896/–753) of the GnSE. The
constructs were then transiently transfected into T3-1 cells in
parallel with the promoterless vector pCAT-basic, which served as a
control for basal levels of CAT activity. Promoter activities were
compared with those of reference constructs containing either the
full-length promoter (–1,135/–26) or a proximal region extending from
–433 to –26 (Fig. 1B). Consistent with our previous studies (21), the
construct containing the full-length 1.1-kb 5'-flanking region
displayed a significantly higher CAT activity than the construct
containing the proximal region only (15.8 ± 2.0- vs.
4.3 ± 1.1-fold increase in CAT expression over the promoterless
vector). However, in contrast with data obtained in the study using the
TK promoter (21), fusion of GnSE with the GnRH-R proximal region
(distal/proximal fusion promoter) resulted in CAT activity equivalent
to that of the full-length promoter (19.1 ± 3.0-fold).
Furthermore, under these circumstances, the 5'-distal region of the
GnSE as well as the 3'-end strongly stimulated activity of the proximal
region (16.2 ± 2.5 and 11.7 ± 1.8-fold, respectively).
Thus, these data suggest that the proximal region, which extends from
–412 to –26, was capable of mediating the positive regulation
exerted by the GnSE, whereas the intermediate domain (–753 to –412)
was unnecessary. In addition, GnSE apparently contained active elements
in both its 5' (–1,135/–900) and 3' (–896/–753) regions.
|
T3–1 cell line as a
gonadotrope-derived cell model. However, although the T3–1 cells
express the gonadotropin -subunit and the GnRH-R gene, unlike normal
gonadotropes no expression of the gonadotropin ß-subunit has been
observed. We, therefore, extended our investigations to the
gonadotrope-derived cell line, LßT2, which has been shown to express
the gonadotropin -subunit and the GnRH-R gene, as well as the
LHß-subunit gene (30, 31) and, under activin treatment, the
FSHß-subunit gene (32). The use of this cell line required an
optimization in the sensitivity of the transfection assay. For this,
the reference constructs were subcloned into the pGL3 basic vector
containing the luciferase reporter gene.
To analyze the relative contribution of the cis-active
elements within the proximal region that could potentially mediate the
positive action of GnSE, we altered the full-length and proximal
promoter by introducing three block-replacement mutations within the
related GRAS, AP-1, or SF-1 elements. These mutant promoters were
compared with the wild-type promoter and reference constructs (Fig. 1C
). As a consequence of the presence of the GnSE, the entire promoter
elicited significantly higher activity than the proximal region alone,
in both T3–1 and LßT2 cells. Mutation of the related GRAS element
(GRASmut) caused a weak or nonsignificant decrease in the transcription
efficacy of the proximal and the full-length promoter in both cell
lines. Alignment of the rat (–412/–395) and mouse (–395/–378) GRAS
sequences revealed a single base pair modification at position –399 (A
to G) that might be responsible for the weak efficacy of the rat
element. Indeed, a similar modification (AA to CC) decreased by 60%
the activity of the mouse GRAS element (24). In contrast,
block-replacement mutation of the related AP-1 site (APmut) markedly
affected luciferase activity, decreasing proximal and full-length
promoter activity by 85% and 68%, respectively, in T3–1 cells.
Similar decreases were observed in LßT2 cells (63% and 40%,
respectively). Nevertheless, AP-1 mutation had no effect on GnSE
activity since the efficiency of the full-length promoter remained
significantly higher than that of the proximal promoter. More
importantly, with regard to the mediation of GnSE activity, mutation of
the SF-1 site (SFmut) displayed stronger efficiency in the full-length
than in the proximal promoter context. Disruption of the SF-1 element
abolished the differences in activity observed between the full-length
and the proximal promoter. This was particularly evident in LßT-2
cells, suggesting that the SF-1 mutation affected both SF-1- and
GnSE-dependent cis-acting efficiencies. The SF-1- and the
AP-1-related elements were thus crucial for gonadotrope-specific
activity of the rat promoter. In addition, SF-1, rather than AP-1,
could mediate the effect of the GnSE.
A 50-bp Sequence, Which Included the SF-1 Element, Was Capable of
Mediating the GnSE Effect
The cell models were used to determine first, if a restricted part
of the proximal region (–275/–226) encompassing the SF-1-related
element could effectively mediate the stimulatory effect of the GnSE
and second, if this activation was gonadotrope specific. To this aim,
we designed three artificial promoters, based on the heterologous
minimal PRL promoter (Fig. 2A). The first
construct contained the GnSE fused to a single copy of the –275/–226
region containing the SF-1 element (SF-1 50 bp module) with both
elements placed upstream of the PRL promoter. Similar constructs
containing a 50-bp module that included either the GRAS (–412/–362)
or AP-1 (–370/–321) element were also generated. Constructs
containing the PRL promoter alone, the 50-bp modules placed upstream of
the PRL promoter, the GnSE fused immediately upstream of the PRL
promoter, or the promoterless vector were tested in parallel by
transient transfection assays in T3–1, LßT2, and Chinese hamster
ovary (CHO) cells.
|
T3–1 and LßT2 cells (Fig. 2). In both cell lines, the relative
luciferase activity of cells transfected with the minimal PRL promoter
was not significantly different (P > 0.05) from that
observed with the promoterless vector. In addition, the GnSE alone
induced a moderate increase in luciferase activity (2- to 3-fold
stimulation over the promoterless vector). In the absence of the GnSE,
the three modules containing either the GRAS-, AP-1-, and SF-1-related
elements induced either no increase (GRAS module in LßT2 cells) or
modest increases in luciferase activity, and maximal stimulation was
observed with the AP-1 module in T3–1 cells (4.1 ± 1.1-fold
stimulation). The promoter activity of the GnSE associated with the
GRAS module was equivalent to that of the GRAS module alone in the two
gonadotrope cell lines. Combination of the AP-1 module with the GnSE
resulted in stimulation of luciferase activity that was approximately
equivalent to the sum of the stimulation induced by the individual
elements in T3–1 and LßT2 cells (6.4 ± 1.1- and 4.3 ±
1.4-fold, respectively). Finally, the association of GnSE with the
–275/–226 region containing the SF-1-related element induced a
dramatic increase in luciferase activity in both T3–1 and LßT2
cells indicative of a synergistic stimulation of the minimal PRL
promoter in these cells (13.0 ± 1.6-fold and 33.1 ±
7.9-fold, respectively). A single-point mutation in the SF-1 element in
this promoter (G to A at position –244, SF*) resulted in a distinct
reduction in promoter activity of 48% and 66% in T3–1 and LßT2
cells, respectively. Complete disruption of the SF-1 element (SFmut)
further decreased the GnSE/SF-1 module efficiency by 65% in T3–1
cells and 86% in LßT2 cells. The GnSE activity, therefore,
required the presence of gonadotrope-specific elements within the
–275/–226 region, in particular the SF-1 element.
Gel-Shift Experiments with Oligonucleotide Probes Overlapping the
–275/–226 Region Revealed a Single Major Complex That Involved the
Potential SF-1 Element
To identify the possible factors that could interact with the
–275/–226 region, gel retardation assays were performed with nuclear
extracts isolated from T3–1 cells (Fig. 3). We used two overlapping
oligonucleotide probes that extended from either –277 to –240 or
–264 to –231. With the labeled –277/–240 probe (Probe-1, Fig. 3C),
no specific complex was detected under the conditions used (data not
shown). In contrast, using the –264/–231 probe (Wild, Fig. 3C), a
major shifted complex was observed, the specificity of which was
confirmed by homologous competition with an excess of unlabeled probe
(10-, 100-, or 1,000-fold molar excess). To localize more precisely the
cis-element involved in the complex formation, three
oligonucleotides spanning the –264/–231 region and containing
8-bp block-replacement mutations at positions –260/ –253 (mut
A), –251/–244 (mut B), and –244/–237 (mut C) were designed. The A,
B, and C mutant oligonucleotides (see Fig. 3C) were then used together
with the labeled wild-type oligonucleotide in competition experiments
using T3–1 nuclear extracts. The A and B mutants were able to
abrogate complex formation in a dose-dependent manner and with a
similar apparent affinity as compared with the unlabeled wild-type
oligonucleotide. In contrast, in the presence of an excess of unlabeled
mutant C, complex formation was unmodified, indicating that the major
shifted complex, obtained with the wild-type oligonucleotide, involved
the sequence extending from –244 to –237, which corresponded to the
potential SF-1 binding site. These results were confirmed using labeled
mutant oligonucleotides (not illustrated). Heterologous competition
using an unlabeled oligonucleotide identical to the sequence containing
the SF-1 element of the rat aromatase gene (SFArom, Fig. 3C) was
performed. Although this oligonucleotide exhibited substantial sequence
differences with the GnRH-R proximal region, complete abrogated complex
formation was obtained at a 100-fold molar excess.
|
T3–1
cells.
The SF-1 Site Was a Determinant for Mediating the Effect
of the GnSE
To investigate the importance of the SF-1 site directly,
the previously described 50-bp modules were replaced by two copies of
the SF-1 site, combining wild-type or mutated elements (Fig. 4A). In addition, overexpression of SF-1
was performed by cotransfecting T3–1, LßT2, and CHO cells with an
SF-1 expression vector. To evaluate potential squelching effects,
control cells were cotransfected in parallel with identical amounts of
cytomegalovirus-ß (CMVß). Under these control conditions and as
expected, none of the constructs displayed greater activity than the
minimal PRL promoter in CHO cells (Fig. 4B, right panel, open
bars). In transfected T3–1 cells, two wild-type SF-1 copies
induced a modest increase in the activity of the minimal PRL promoter
but failed to mediate the effect of the GnSE (Fig. 4B, left
panel, open bars). In contrast, in control LßT2 cells, the
duplicated wild-type SF-1 sites were able to mediate the effect of the
GnSE, and the level of promoter activity attained 5.2 ± 0.5-fold
that of the minimal PRL promoter (Fig. 4B, middle panel, open
bars). Disruption of one or two SF-1 elements drastically reduced
the promoter activity and resulted in a construct that displayed the
same activity as the minimal PRL promoter, suggesting that the SF-1
mutations concomitantly abolished SF-1- and GnSE-dependent promoter
activity in LßT2 cells.
|
). The SF-1 element
of the GnRH-R promoter was thus functional and sufficient for mediating
SF-1-dependent transactivation in gonadotrope-derived as well as in CHO
cells. When SF-1 was overexpressed, GnSE significantly enhanced the
activity of the promoter containing the wild-type SF-1 elements in
T3–1 and LßT2 cells (3.5 ± 0.4 vs. 2.3 ±
0.2 and 8.4 ± 0.5 vs. 4.9 ± 0.6-fold over the
minimal PRL promoter vector, respectively) but not in CHO cells
(2.4 ± 0.2 vs. 3.3 ± 0.3). Disruption of the two
SF-1 elements abrogated both GnSE- and SF-1-dependent transactivations.
Overexpression of SF-1 thus increased the efficiency of the GnSE in
LßT2 cells and was sufficient to restore the capacity of the SF-1
element to mediate GnSE activity in T3–1 cells. However, the
isolated SF-1 elements displayed lower efficiency than the proximal
50-bp region in mediating GnSE activity. The sequence surrounding the
SF-1 element might be necessary for promoting the efficient recruitment
of SF-1 to its cognate element, thereby facilitating binding and
transactivation. In contrast, the GnSE remained inactive in CHO cells
overexpressing SF-1 despite the potency of the SF-1 elements to
transactivate the minimal PRL promoter implying that other factors,
absent in these cells, could be required for interaction with the
GnSE.
The Potency of the GnSE Resided within Two Separate Regions of
Approximately 30 bp
In an attempt to localize the active elements of the GnSE,
serial 5'- and 3'-deletion mutants of the full-length GnSE were placed
upstream of an artificial promoter containing duplicated copies of the
SF-1 50-bp module fused to the PRL promoter. This artificial promoter
was able to mediate GnSE activity 3 to 7 times as much as the previous
promoter containing a single SF-1 module (see Fig. 2) and thus was used
to ensure accuracy in the analysis. As illustrated in Fig. 5, deletions that extended from –1135 to
–1063 were inefficient in both T3–1 and LßT2 cells
(P > 0.01). In contrast, 5'-deletions in a 114-bp
region situated between –1063 and –950 severely affected the
efficiency of the GnSE in T3–1 cells. Further deletion of a 71-bp
region extending from –896 to –826 resulted in a promoter activity
equivalent to that of the duplicated 50-bp module (2.4 ± 0.3
vs. 2.8 ± 1.5-fold over the promoterless vector,
P > 0.05). The importance of the 114- and 71-bp
sequences was further attested by independent results obtained with
5'-/3'-deleted constructs. In fact, 3'-deletion of the –900/–753
region, which included the 71-bp sequence identified above, yielded a
substantial reduction (
60%) in the efficiency of the GnSE as
compared with the full-length construct (8.8 ± 0.8 vs.
20.0 ± 1.1-fold over the promoterless vector, P
< 0.001). Additional 5'-deletion of the 114-bp region completely
abrogated GnSE activity, resulting in promoter activity that was
equivalent to that of the duplicated 50-bp module.
|
T3–1 cell line.
The 114-bp region accounted for approximately 45% (range 35%–60%)
of the overall efficiency of the GnSE in LßT2 cells as compared with
70% (range 66%–75%) in T3–1 cells. On the other hand, the
decrease in promoter efficiency induced by a deletion of the
–896/–826 region was more obvious in LßT2 than in T3–1 cells.
The 71-bp region accounted for 55% (range 43%–63%) as compared with
30% (range 25%–34%) in the respective cell lines.
Additional 5'- and 3'-deletions were created in the context of the
artificial promoter and T3–1 cells were transfected with the
resulting constructs. The active elements of the 114-bp sequence
(5'GnSE) were further delimited to a region extending from –983 to
–950 whereas those of the 71-bp sequence (3'GnSE) were found to be
located between –896 and –859 (data not illustrated).
Block-Replacement Mutagenesis within the Wild-Type Promoter
Confirmed the Bipartite Organization of the GnSE
These active regions were then scanned by adjacent 10-bp
block-replacement mutations in the wild-type promoter, and the mutant
constructs were tested by transient transfection in both gonadotrope
cell lines (Fig. 6A). Wild-type
full-length and proximal promoter constructs were transfected in
parallel. In T3–1 cells, mutations extending from –983 to –974
(GA) and from –971 to –962 (GB) altered significantly promoter
activity, inducing a 45–50% decrease in the stimulation induced by
GnSE (Fig. 6B). This suggested that the DNA sequence overlapped by
these two mutations contained positive regulatory elements that were
most likely responsible for the activity of the 5'-region of GnSE.
Among the three block-replacement mutations that altered the 3'-region
of GnSE only the GF mutation (–871 to –862) was proficient and
reduced GnSE efficiency by 37%. Similar results were obtained after
transfection of LßT2 cells: the GA, GB and GF mutations
decreased the GnSE-induced stimulation by 47%, 63%, and 62%,
respectively. In addition, the GD mutation (–895/–886), which was
ineffective in T3–1 cells, induced a 47% decrease in LßT2
transfected cells.
|
).
In addition, two GATA elements were present in the sense and antisense
strand, one of which corresponded to the canonical motif WGATAR. The
sequence altered by mutation GF harbored a palindrome element
containing consensus half-sites that are recognized by the
LIM-homeodomain factors Lhx2 (also designated LH-2), Isl I, and Lmx I
(33). Finally, the sequence overlaid by mutation GD encompassed an
upstream stimulatory factor (USF) element that matched perfectly with
the consensus binding site (CACGTG). However, since GD mutation
appeared to affect promoter activity in LßT2 cells only while GnSE
was active in both gonadotrope cell lines, we restricted further
analyses to GnSE elements altered by mutation GA, GB, and GF.
Gel-Shift Experiments with Oligonucleotide Probes Overlapping the
GA/GB Region Revealed the Potential Implication of GATA-Related Factors
in GnSE Activity
To examine whether the GA/GB region could bind specific
transcription factors, we designed a double-stranded oligonucleotide
(AB probe) which extended from –985 to –957 and gel retardation
assays were performed with nuclear extracts isolated from both T3–1
and LßT2 cell lines. Using a labeled AB probe, two distinct shifted
complexes (complexes 1 and 2) were observed in T3–1 and LßT2
nuclear extracts, and the specificity was confirmed by homologous
competition with an excess unlabeled probe (5-, 10-, 50-, and 100-fold
molar excess) (Fig. 7A). Depending on the
gel resolution, the faster-migrating complex (complex 2) was actually a
composite of two distinct complexes (Fig. 7A, right panel,
and Fig. 7, panels B and C). We then designed two mutant
double-stranded oligonucleotides that differed from the wild-type AB
probe by 3 bp located at position –977 to –975 (mut1) and at position
–970 to –968 (mut 2). These mutations disrupted either the NFY
binding site or the GATA antisense and Ptx1 binding site, respectively.
The mut1 and mut2 oligonucleotides were then used together with the
labeled AB probe in competition experiments. The mut1 oligonucleotide
with the disrupted NFY binding site was able to abrogate complex
formation in a dose-dependent manner and with a similar apparent
affinity as compared with the unlabeled wild-type oligonucleotide,
suggesting that NFY factor was not involved in complex formation (not
illustrated). In contrast, the mut2 oligonucleotide displayed binding
capacity that differed from wild-type AB probe. Only the
faster-migrating complex (complex 2) was detected (Fig. 7B), indicating
that the –970/ –968 mutation within the GATA antisense/Ptx-1
binding site abrogated complex 1 formation. These data were confirmed
by using labeled AB probe and unlabeled mut2 oligonucleotide in
competition experiments (not illustrated).
|
-globin promoter. In parallel, we also tested
the capacity of a NFY oligonucleotide, derived from the rat albumin
promoter to abrogate complex formation. As expected and consistent with
the results presented above, complex formation was unmodified in the
presence of an excess of unlabeled NFY oligonucleotide (not shown).
Interestingly, the POMC Ptx-1 oligonucleotide was unable to abrogate
the shifted complexes obtained with either the labeled AB or the mut2
probe (Fig. 7, B and C). In contrast, the binding of the AB and mut2
probe was clearly competed by a 100-fold molar excess of the GATA
oligonucleotide (Fig. 7, B and C), suggesting that GATA-related factors
interacted with the 5'-region of GnSE. Furthermore, since a mutation at
position –970 to –968 abrogated the shifted complex 1, the antisense
GATA element (–965/–969) was likely involved in its formation.
Consequently, since neither mutation at position –977 to –975 nor at
position –970 to –968 affected the faster-migrating complex (complex
2), it could involve the canonical GATA element in the sense
orientation. Similar results were obtained with T3–1 nuclear
extracts. Altogether these findings suggest that the AB probe contained
two binding sites for GATA-related factors that might account for the
activity of the 5'-region of GnSE.
The Binding Activity of the 3'-Region of GnSE Was Abolished by
Mutation of the Palindrome LIM-Homeodomain Element
To determine whether the sequence overlaid by mutation GF could
also bind specific factors, a labeled double-stranded oligonucleotide
(F) corresponding to sequence –880 to –852 of GnSE was used as a
probe with nuclear extracts prepared from T3–1 and LßT2 cells. A
single shifted complex was detected, which was competed with a 100-fold
molar excess of the unlabeled oligonucleotide F but not by a
double-stranded oligonucleotide bearing the GF mutation (Fig. 8). In addition, this DNA binding
activity was not competed by heterologous GATA or Ptx1 probes. Enhancer
activity of the 3'-region of GnSE thus correlated with a binding
activity that was different from that in the 5'-region of GnSE.
|
) and
transactivation (see Fig. 6). Among the members of the LIM- homeodomain
family of transcription factors, Lhx2 was shown to be expressed in
T3–1 cells and to stimulate constitutive transcription and
hormonally regulated expression of the mouse gonadotropin -subunit
gene (33). Lhx2 could bind to a pituitary-specific enhancer designated
the pituitary glycoprotein basal element (PGBE) which is present in the
-subunit gene promoter. Alignment of the –871/–859 sequence of the
GnSE with the –350/–323 sequence of the PGBE indicated that the
perfect palindrome CTAATTAG of GnSE was strongly homologous to both the
upstream and downstream region of the binding site of Lhx2 within the
PGBE (Fig. 9A). We, therefore, evaluated
the ability of a PGBE oligonucleotide (Fig. 9B) to abrogate the
specific complex obtained with the labeled F probe. A 50- and 100-fold
molar excess of unlabeled PGBE clearly decreased the abundance of the
specific complex, suggesting that the palindrome sequence of the GnSE
interacted with protein(s) that were related to PGBE binding factor
such as Lhx2. Mutations in PGBE that were previously shown to abolish
binding of Lhx2 (33) also abrogated the ability of the heterologous
oligonucleotide to compete with F probe binding (Fig. 9B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-subunit, the LHß-subunit, and the
GnRH-R (see review in Ref. 39). It has been shown, using transient
transfection assays in T3–1 cells that, in addition to SF-1,
constitutive expression of the GnRH-R gene requires AP-1 and, an as yet
unknown, GRAS-binding protein. These factors interact with a tripartite
basal enhancer located in the proximal region of the promoter (Refs 23, 24 ; see Introduction). In the present study, we have
shown that the cell-specific expression of the rat GnRH-R gene is also
dependent on the activity of two novel positive regulatory regions in
GnSE localized further upstream of the tripartite enhancer.
The GnSE exhibits an attractive feature that is likely related to
mechanisms underlying the expression of the GnRH-R gene in cells of the
gonadotrope lineage. Convergent data in fact indicate that the SF-1
response element is crucial for mediating the effect of GnSE. First,
mutations that disrupt the putative SF-1 response element diminished,
most effectively, the activity of the full-length promoter, as compared
with mutations in GRAS or AP-1 elements. This was particularly evident
in LßT2 cells. Second, the GnSE could cooperate in a synergistic
manner with a limited part of the proximal promoter region that
includes the putative SF-1 binding site but not with modules of similar
length that contain either GRAS or AP-1 element. In this proximal
region, a single specific protein/DNA complex detected in gel-shift
experiments was shown to implicate the SF-1 response element and a
factor that was closely related to SF-1. Finally, in SF-1
overexpressing T3–1 and LßT2 cells, two SF-1 elements were able
to mediate the effect of GnSE, and mutations in SF-1 elements
completely abrogated this capacity.
Altogether, these data demonstrate the existence of a functional
interaction between the GnSE and the SF-1 response element, suggesting
possible protein/protein interactions with transcription factors that
bind to GnSE. Block-replacement mutations within the 5'- and 3'-region
of GnSE combined with gel retardation experiments suggest the
implication of GATA and LIM homeodomain-related factors. Proteins such
as Lhx2 belonging to the LIM family as well as GATA factors are present
in gonadotrope cells and have been shown to participate in the
constitutive and/or GnRH-regulated expression of the gonadotropin
-subunit in T3–1 cells (33, 40). However, a physical interaction
between Lhx2 and SF-1 has not yet been demonstrated, although an
increasing list of transcription factors have been shown to have a
functional cooperation with SF-1 including stimulating protein 1 (Sp1)
(41), CCAAT/enhancer-binding protein ß (C/EBPß) (42), Wilm’s tumor
1 (WT-1) (43), SRY-box containing gene 9 (SOX9) (44), and c- JUN (45).
Cooperation with SF-1 was also observed with GATA-4, resulting in a
synergistic stimulation of the antimullerian hormone/mullerian
inhibiting substance (AMH/MIS) promoter (46). Furthermore, several
factors belonging to the GATA family of transcription factors display
similar capacity. In addition, GATA-2 appears to be decisive in the
early stages of gonadotrope cell differentiation. In transgenic mice
that express GATA-2 under the control of the Pit-1 promoter, the
gonadotrope population was dramatically expanded, reaching 90% instead
of the usual 10% of the total pituitary cell population, with a
concomitant failure in the differentiation of lactotropes,
somatotropes, and, to a minor extent, thyrotropes (34). This suggests
that GATA-2 controls the expression of several marker genes of the
gonadotrope lineage such as the GnRH-R gene. Our data are, therefore,
consistent with a model in which members of the GATA family mediate
GnSE activity in cooperation with SF-1 as suggested in the present
study by gel retardation experiments.
It was previously reported that promoter activity of the GnRH-R gene is
enhanced in transient transfection assays by cotransfection of a Ptx1
expression vector and, correspondingly, the expression of the mouse
GnRH-R gene is decreased in Ptx1 knock-down T3–1 cells (47).
However, despite the presence of a consensus binding site on the sense
strand at position –971/–966, Ptx1 does not appear to be implicated
in GnSE activity. Nevertheless, it may regulate GnRH-R gene by
interacting with other potential Ptx1 binding sites or by modulating
SF-1 activity through direct protein-protein interaction. Indeed,
synergistic activation of transcription by Ptx1 and SF-1 of the
LHß-subunit promoter is not prevented by disruption of the Ptx1
binding site, indicating that Ptx1 can also cooperate with SF-1 through
DNA-independent interaction (48).
In the case of the LHß-subunit and AMH/MIS promoters, the binding sites for SF-1 and interacting transcription factors are close to each other. Binding sites for SF-1, Ptx1, and Egr-1 are indeed clustered within approximately 80 bp in the proximal region of the LHß-subunit promoter (49). Likewise, in the proximal AMH/MIS promoter, the SF-1 and GATA response elements are 10 bp apart (46). The situation is very different in the GnRH-R promoter since the GnSE upstream element is separated from the SF-1 site by approximately 700 bp in the wild-type promoter. Thus, the GnSE belongs to a class of promoter-specific enhancers, capable of acting from a distance as classical enhancers, but requiring a specific promoter context.
In addition to its ability to interact with several transcription
factors, SF-1 is able to bind at least two coactivators, the cAMP
regulatory element-binding protein (CREB) binding protein (CBP/p300)
(50) and the steroid receptor coactivator 1 (SRC-1) (51, 52) and to
recruit the general transcription factor TFIIB (44, 53). SRC-1 is known
to interact with several nuclear hormone receptors and augment
ligand-dependent transactivation. SRC-1 is also capable of interacting
with CBP/p300 via a separate domain (reviewed in Ref. 54). In addition,
our data demonstrated that, in the context of the minimal PRL promoter,
SF-1 in the absence of any other additional element enhances
transcription (Fig. 4). SF-1 is thus likely capable of recruiting the
general transcription factors and the RNA polymerase II through direct
interaction with CBP/p300 or TFIIB or indirectly through SRC-1. In
contrast, GnSE behaves differently since it requires the presence of
proximal elements such as those located in the 50-bp proximal region,
namely SF-1, to activate the minimal PRL promoter in a synergistic
manner. This is reminiscent of the differential properties of the
coactivator CBP/p300 as compared with its associated factor pCAF
(p300/CBP associated factor) in their capacity to activate
transcription (55). To be efficient, CBP/p300 must be recruited to the
vicinity of the core promoter region whereas pCAF activates
transcription from a pro-moter-distant position provided an
upstream activator element is present in the vicinity of the core
promoter region. Similarly, it might be hypothesized that GATA- and/or
Lhx2-related factors that bind the GnSE recruit a co-activator such as,
or with similar properties as, pCAF. This putative coactivator would
subsequently interact with CBP recruited to the vicinity of the core
promoter region through SF-1/SRC-1 interactions. Recently, evidence has
been obtained that the melanocyte-specific gene-related gene 1 (MRG1)
may bind Lhx2 in vitro and form a complex with Lhx2 in
T3–1 cells (56). Furthermore, MRG1 is also able to bind CBP/p300.
Therefore, the functional link that occurs between the GnSE and the
GnRH-R proximal promoter region, which includes the SF-1 response
element, might be established through coactivator interaction.
Although GnRH-R promoter fusion constructs as well as artificial
promoter constructs displayed rather similar patterns of activity after
transfection in gonadotrope-derived cell lines, differences were
consistently observed between LßT2 and T3–1 cells. The proximal
–412/–26 region as well as the duplicated copies of the
GnRH-R-specific SF-1 element are significantly more efficient in LßT2
than in T3–1 cells (Figs. 1 and 4). Furthermore, mutations in SF-1
element affected, to a larger degree, the activity of the artificial
promoter in LßT2 than in T3–1 cells (Fig. 2). Such differences
may result from differential representation of transcription factors,
especially SF-1, which, in turn, may be related to the way in which the
two cell lines have been selected (29). The T3–1 cells were derived
from pituitary tumors induced in transgenic mice by directed
oncogenesis with the simian virus 40 T antigen under the control of the
human gonadotropin -subunit promoter. They express the gonadotropin
-subunit and GnRH-R genes. The LßT2 cells were obtained by a
similar approach except that the rat LHß-subunit promoter was used
for directing the expression of the oncogene (30, 31). These cells
express not only the gonadotropin -subunit and GnRH-R genes but also
the LHß- and, under activin treatment (32), the FSHß-subunit genes.
Interestingly, the gonadotropin -subunit and the GnRH-R genes are
expressed in the early stages of pituitary ontogenesis whereas the
ß-subunits of LH and FSH are expressed in the late stages (Refs.
57, 58, 59, 60, 61, 62 ; also reviewed in Refs. 34, 63). It could be thus suggested
that the T3–1 cells are representative of immature gonadotrope
cells whereas LßT2 cells are almost fully differentiated gonadotrope
cells. Our results indicate that the SF-1 element alone is capable of
mediating the effect of the GnSE in LßT2 cells whereas it appears
insufficient in T3–1 cells. In this cell line, overexpression of
SF-1 is necessary to detect the enhancer capacity of the GnSE. Based on
these findings, it may be hypothesized that the SF-1-dependent
regulation of the GnRH-R gene is correlated with the stage of
gonadotrope differentiation, playing a more significant role in
differentiated than in immature gonadotrope cells. This may be achieved
through variations in the intracellular concentration of SF-1 as well
as through posttranslational modifications such as phosphorylation (64)
that might affect SF-1 transactivation efficacy. This is more
consistent with data showing that the expression of the GnRH-R gene is
not strictly dependent on SF-1. Indeed, the gonadotrope lineage is not
ablated after targeted disruption of the gene encoding SF-1 in mice
since treatment with GnRH restores gonadotrope function (65).
In T3–1 cells, the 50-bp region is efficient for mediating
the effect of the GnSE and we thus propose that it promotes the
recruitment of SF-1 to its binding site. Our observation that
overexpression of SF-1 restored the capacity of the SF-1 element to
mediate the activity of GnSE is consistent with this hypothesis. An
alternative but nonexclusive possibility could be that additional
transcription factors interact with the 50- bp region and mediate in
place of, and/or in cooperation with, SF-1 the effect of the GnSE.
These putative factors, however, were not detected in gel-shift
experiments, suggesting that they interact weakly with their cognate
DNA response elements. Nevertheless, they would provide a means by
which, ultimately, the GnSE-bound factors may bypass the deficiency in
SF-1, either under pathological conditions such as those artificially
created in SF-1-disrupted mice, or under physiological conditions,
e.g. during the early stages of pituitary ontogenesis that
precede the appearance of SF-1 in the pituitary.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
or illustrated in Figs. 3, 7, 8, and 9. The nucleotide sequence of the
GnRH-R gene is numbered relative to the translational initiation codon
[GenBank accession numbers Z99955 and Z99956 (21)]. The rat PRL gene
is numbered relative to the transcription start site (66). All PCR
amplifications were performed with either Deep Vent DNA polymerase
(New England Biolabs, Inc., Montigny-le Bretonneux,
France) or Expand High Fidelity PCR System (Roche Molecular Biochemicals, Meylan, France).
|
To subclone the GnRH-R promoter fragments upstream of the luciferase reporter gene into pGL3-Basic vector (Promega Corp.), the multiple cloning site of the pGL3-Basic vector was modified to provide compatible restriction sites in the appropriate orientation. The selected sense and antisense oligonucleotides MCS-S and MCS-A were annealed, and the resulting double-stranded oligonucleotide with 5'- and 3'-protruding ends was inserted into the KpnI/HindIII sites of the pGL3-Basic vector. The multiple cloning site of the modified pGL3-Basic vector thus included, in the 5'- to 3'-direction, restriction sites for HindIII, KpnI, BstEII, and XhoI. Full-length and proximal promoter fragment were generated by PCR amplification from GnRH-R promoter template using sense primers –1,135Hind and –475BstE, respectively, and antisense –26Xho, gel purified, and subcloned into the HindIII/XhoI or BstEII/XhoI sites of the modified pGL3-Basic vector. The GRAS, AP-1, and SF-1 mutant constructs were prepared following the same protocol used for insertion of the KpnI site into the proximal promoter construct. The GRAS, AP-1, and SF-1 elements were modified so that a restriction site for KpnI was created using mutagenic sense primers GRAS-MutS, AP1-MutS, or SF1-MutS and antisense primer Luc-A in one PCR and using sense primer Luc-S and mutagenic antisense primers GRAS-MutA, AP1-MutA, or SF1-MutA in another PCR; both PCR reactions were performed with the pGL3 plasmid containing the full-length GnRH-R promoter as a template. Overlapping PCR products were then self annealed, and DNA double-strands were reconstituted in a third PCR using sense primer Luc-S and antisense primer Luc-A.
The artificial promoter constructs were generated in successive steps. First, the minimal PRL promoter (–35/+36) was amplified by PCR from the p0.4 kb PRL-CAT vector (67) using sense primer BstE-S and antisense primer Xho-A, digested with XhoI and BstEII restriction enzymes, gel purified, and inserted into the XhoI-BstEII site of the modified pGL3-Basic vector. Second, single 50-bp modules containing the GRAS, AP-1, or SF-1 elements were amplified using sense primer –412Kpn-S, –370Kpn-S, or –275Kpn-S, respectively, and antisense primer –362BstE-A, –315BstE-A, or –229BstE-A, respectively, and processed as above to generate constructs containing a single copy of the –412/–362, –370/–321, or –275/–226 region, respectively. Two copies of the 50-bp module containing the SF-1 site were generated from the GnRH-R promoter template in a separate PCR using sense primer –275Bam-S or –275Kpn-S and antisense primer –229BstE-A or –226Bam-S, respectively. These were digested by restriction enzymes corresponding to the sites introduced by the primers and gel purified. Two 50-bp modules were created containing either 5'-KpnI and 3'-BamHI or 5'-BamHI and 3'-BstEII cohesive ends inserted together at the KpnI-BstEII site into the modified pGL3 vector containing the minimal PRL promoter. DNA fragments containing two copies of the wild-type or mutated SF-1 elements (–245/–237) were obtained by the hybridization of SFx2S and SFx2A complementary oligonucleotides, SFMut1/2S and SFMut1/2A or SFMut2/2S and SFMut2/2A. The resulting double-stranded DNA fragments with 5'-BstEII recessing and 3'-KpnI recessing ends were then subcloned in place of the 50-bp region. Finally, the full-length GnSE as well as the 5'- and/or 3'-deleted constructs were generated by amplification using selected sense primers –1,135Hind, –1,063Hind, –950Hind, –896Hind, or –826Hind and antisense primers –753Kpn or –900Kpn. The resulting amplified products containing the desired 5'- and/or 3'-deletions were then digested with HindIII and KpnI restriction enzymes and inserted into the HindIII-KpnI site of the modified pGL3-Basic vector containing the PRL promoter and the 50-bp modules. Block-replacement mutation (GA to GF) within the upstream and downstream region of GnSE were introduced after the same three-step PCR protocol as that used for generating previous mutants. The sense primers GA-S to GF-S were combined with antisense primer Luc-A, and the antisense primers GA-A to GF-A were combined with Luc-S. The identity of all reporter constructs was confirmed by sequencing using the dideoxynucleotide chain termination method.
Cell Culture and Transient Transfection
LßT2 and T3–1 cells were maintained in monolayer cultures
in high glucose DMEM supplemented with 10% FBS, 100 U/ml penicillin,
and 100 µg/ml streptomycin sulfate (Sigma, St. Louis,
MO) at 37 C in humidified 5% CO2/95% air. CHO
cells were cultured as described (68). For transient transfection
experiments with CAT reporter constructs, cells were plated at 8
x 105 cells per 60-mm dish the day before
transfection. On the day of transfection, equivalent molar amounts of
reporter constructs were combined with 4 µl PLUS reagent and 6 µl
lipofectAMINE reagent (Life Technologies, Inc.,
Gaithersburg, MD) in 500 µl OptiMEM medium (Life Technologies, Inc.) according to manufacturer’s instructions. In each
experiment the total quantity of DNA per dish was standardized to 3
µg with pUC19 plasmid DNA. A vector expressing ß-galactosidase
under the control of the CMV promoter (0.4 µg per dish) (CMVß ,
CLONTECH Laboratories, Inc. Palo Alto, CA) was
cotransfected to serve as an internal standard for transfection
efficiency. Transfection mixture was incubated for 30 min at room
temperature, diluted to 2.5 ml with OptiMEM medium, and applied to the
cells previously washed with the same serum-free medium. After a 6
h-incubation, the medium was replaced by DMEM supplemented with 2%
FBS, 10 U/ml penicillin, and 10 µg/ml streptomycin sulfate, and
16 h later, the medium was aspirated and cells were processed as
previously described for ß-galactosidase and CAT assays (21).
For transient transfection experiments with luciferase reporter constructs, cells were plated at 105 cells per well in 24-well plates. They were then processed according to the same protocol as for CAT reporter constructs, except that all quantities and volumes were scaled down 10-fold and Renilla luciferase expression vector (pRL-TK, 10 ng/dish) was used as an internal control to normalize gene expression measurements instead of CMVß. Firefly and Renilla luciferase activities were then measured using the Dual-Luciferase Reporter Assay System (Promega Corp.). Cells were washed once with PBS, pH 7.4, and lysed by the addition of 100 µl passive lysis buffer. Firefly luciferase activity was assayed using 1–4 µl cell extract combined with 5–20 µl luciferase assay reagent, and luminescence was measured with a 15-sec delay and a 15-sec measurement in a Wallach scintillation counter from which the coincidence circuit was turned off. An equal volume of Stop & Glo reagent was then added, and Renilla luciferase activity was determined under the same conditions.
EMSA
Cells were seeded at 3 x 106
(T3–1) or 6 x 106 cells (LßT2) per
100-mm tissue culture dish in triplicate and cultured for 24 h in
DMEM supplemented with 10% FBS, 100 U/ml penicillin, and 100 µg/ml
streptomycin sulfate. The culture medium was then replaced by OptiMEM
medium and cells were cultured for an additional 6 h. The
serum-free medium was then replaced by DMEM supplemented with 2% FBS,
10 U/ml penicillin, and 10 µg/ml streptomycin sulfate, and incubation
was continued for a further 16 h. Thereafter, cells were harvested
and nuclear extracts were prepared by the method of Andrews and Faller
(69).
Double-stranded oligonucleotides were designed to contain 5'-protruding
ends (see Table 1 and Figs. 3, 7, and 8). They were thus end labeled (5
fmol) by filling-in the recessed 3'-termini with Klenow fragment of
Escherichia coli DNA polymerase I and 50 µCi
[32P] dCTP (3000 Ci/mmol, Amersham Pharmacia Biotech, Les Ulis, France), and purified with a
Sephadex G50 fine column.
Nuclear extracts (9 µg) and polydIdC (1 µg) were incubated in
binding buffer [20 mM HEPES, pH 7.9, 60 mM
KCl, 60 mM NaCl, 1 mM EDTA, 300 µg/ml BSA,
and 12% (vol/vol) glycerol] for 15 min at 4C. Thereafter, 20,000 to
50,000 cpm of oligonucleotide probe (
10 fmol) were added with or
without an excess of unlabeled oligonucleotide, and incubation was
continued at 20 C for 30 min. Protein DNA complexes were resolved on a
5% nondenaturing polyacrylamide gel in 0.5x Tris-borate-EDTA buffer.
Gels were then dried and subjected to autoradiography.
Statistical Analysis
Data were analyzed by one-way ANOVA. If the F test was
significant, then means were compared using Tukey-Kramer’s method of
multiple comparisons.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This work was supported by grants from the CNRS and the Université Pierre et Marie Curie. H. Pincas is a recipient of the Ministère de l’Education Nationale, de la Recherche et de la Technologie, and of the Fondation pour la Recherche Médicale.
Received for publication February 21, 2000. Revision received September 22, 2000. Accepted for publication October 23, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-chain
messenger ribonucleic acids that are regulated by GnRH in vitro. J
Clin Endocrinol Metab 84:743–750-subunit gene in pituitary gonadotropes. Mol Endocrinol 8:878–885 subunit promoter by a
LIM-homeodomain transcription factor. Mol Cell Biol 14:2985–2993.-subunit gene in the placenta and
pituitary gland. Mol Cell Biol 14:5592–5602-subunit gene expression. J Biol Chem 274:36159–36167-subunit in the
pars tuberalis of the rat pituitary gland during ontogenesis.
Neuroendocrinology 58:616–624[Medline]
This article has been cited by other articles:
 |
|
 |
|
  R. S. Viger, S. M. Guittot, M. Anttonen, D. B. Wilson, and M. Heikinheimo Role of the GATA Family of Transcription Factors in Endocrine Development, Function, and Disease Mol. Endocrinol., April 1, 2008; 22(4): 781 - 798. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Shadley, K. Divakaran, K. Munson, R. N. Hines, K. Douglas, and D. G. McCarver Identification and Functional Analysis of a Novel Human CYP2E1 Far Upstream Enhancer Mol. Pharmacol., June 1, 2007; 71(6): 1630 - 1639. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Zheng, J. Yang, Q. Jiang, Z. He, and L. M Halvorson Liver receptor homologue-1 regulates gonadotrope function J. Mol. Endocrinol., February 1, 2007; 38(2): 207 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Granger, C. Bleux, M.-L. Kottler, S. J. Rhodes, R. Counis, and J.-N. Laverriere The LIM-Homeodomain Proteins Isl-1 and Lhx3 Act with Steroidogenic Factor 1 to Enhance Gonadotrope-Specific Activity of the Gonadotropin-Releasing Hormone Receptor Gene Promoter Mol. Endocrinol., September 1, 2006; 20(9): 2093 - 2108. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Galmiche, N. Richard, S. Corvaisier, and M.-L. Kottler The Expression of Aromatase in Gonadotropes Is Regulated by Estradiol and Gonadotropin-Releasing Hormone in a Manner that Differs from the Regulation of Luteinizing Hormone Endocrinology, September 1, 2006; 147(9): 4234 - 4244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. P. S. Bhangoo, C. S. Hunter, J. J. Savage, H. Anhalt, S. Pavlakis, E. C. Walvoord, S. Ten, and S. J. Rhodes A Novel LHX3 Mutation Presenting as Combined Pituitary Hormonal Deficiency J. Clin. Endocrinol. Metab., March 1, 2006; 91(3): 747 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. C. Yaden, M. Garcia III, T. P. L. Smith, and S. J. Rhodes Two Promoters Mediate Transcription from the Human LHX3 Gene: Involvement of Nuclear Factor I and Specificity Protein 1 Endocrinology, January 1, 2006; 147(1): 324 - 337. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. McGillivray, J. S. Bailey, R. Ramezani, B. J. Kirkwood, and P. L. Mellon Mouse GnRH Receptor Gene Expression Is Mediated by the LHX3 Homeodomain Protein Endocrinology, May 1, 2005; 146(5): 2180 - 2185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. D. Cherrington, T. A. Farmerie, C. A. Lents, J. D. Cantlon, M. S. Roberson, and C. M. Clay Activin Responsiveness of the Murine Gonadotropin-Releasing Hormone Receptor Gene Is Mediated by a Composite Enhancer Containing Spatially Distinct Regulatory Elements Mol. Endocrinol., April 1, 2005; 19(4): 898 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Granger, V. Ngo-Muller, C. Bleux, C. Guigon, H. Pincas, S. Magre, D. Daegelen, A. Tixier-Vidal, R. Counis, and J.-N. Laverriere The Promoter of the Rat Gonadotropin-Releasing Hormone Receptor Gene Directs the Expression of the Human Placental Alkaline Phosphatase Reporter Gene in Gonadotrope Cells in the Anterior Pituitary Gland as well as in Multiple Extrapituitary Tissues Endocrinology, February 1, 2004; 145(2): 983 - 993. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A. LaVoie The Role of GATA in Mammalian Reproduction Experimental Biology and Medicine, December 1, 2003; 228(11): 1282 - 1290. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. K. Bachir, G. Garrel, A. Lozach, J.-N. Laverriere, and R. Counis The Rat Pituitary Promoter of the Neuronal Nitric Oxide Synthase Gene Contains an Sp1-, LIM Homeodomain-Dependent Enhancer and a Distinct Bipartite Gonadotropin-Releasing Hormone-Responsive Region Endocrinology, September 1, 2003; 144(9): 3995 - 4007. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Garrel, A. Lozach, L. K. Bachir, J.-N. Laverriere, and R. Counis Pituitary Adenylate Cyclase-activating Polypeptide Stimulates Nitric-oxide Synthase Type I Expression and Potentiates the cGMP Response to Gonadotropin-releasing Hormone of Rat Pituitary Gonadotrophs J. Biol. Chem., November 22, 2002; 277(48): 46391 - 46401. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Fowkes, P. King, and J. M. Burrin Regulation of Human Glycoprotein Hormone {alpha}-Subunit Gene Transcription in L{beta}T2 Gonadotropes by Protein Kinase C and Extracellular Signal-Regulated Kinase 1/2 Biol Reprod, September 1, 2002; 67(3): 725 - 734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. K. Bachir, J.-N. Laverriere, and R. Counis Isolation and Characterization of a Rat Nitric Oxide Synthase Type I Gene Promoter that Confers Expression and Regulation in Pituitary Gonadotrope Cells Endocrinology, November 1, 2001; 142(11): 4631 - 4642. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Pincas, J.-N. Laverriere, and R. Counis Pituitary Adenylate Cyclase-activating Polypeptide and Cyclic Adenosine 3',5'-Monophosphate Stimulate the Promoter Activity of the Rat Gonadotropin-releasing Hormone Receptor Gene via a Bipartite Response Element in Gonadotrope-derived Cells J. Biol. Chem., June 22, 2001; 276(26): 23562 - 23571. [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 |
