| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Messenger Ribonucleic Acid Expression of Term Villous Trophoblasts1
Department of Obstetrics and Gynecology, University of Vienna, Währinger Gürtel 18–20, A-1090 Vienna, Austria
Address all correspondence and requests for reprints to: Martin Knöfler, Department of Obstetrics and Gynecology, University of Vienna, Waehringer Guertel 18–20, A-1090 Vienna. E-mail: martin. knoefler{at}akh-wien.ac.at
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß human CG subunits,
which is predominantly governed at the level of messenger RNA
expression. Here, we present a detailed study on the mechanisms
involved in the differentiation-dependent regulation of the
trophoblast-specific CG gene promoter. Site-directed mutations in
each of the five DNA-elements of the composite enhancer were performed
to investigate the contribution of the individual regulatory sequences
to the overall transcriptional activity of the promoter at two
different stages of trophoblast in vitro
differentiation. We show that deletion of one cyclic AMP response
element (CRE) did not affect CG promoter activity in
cytotrophoblasts; however, it reduced transcription by 33% in
differentiating cultures. Removal of both CREs almost abolished
transcription at early and later stages of in vitro
differentiation. Upon mutation the enhancer elements ACT, JRE, and
CCAAT significantly decreased luciferase reporter transcription;
however their contribution to the total promoter activity did not
change during in vitro differentiation. Contrary to
that, mutated TSE diminished promoter activity by 19% during 12 and
48 h of cultivation but reduced luciferase expression by 78%
between 48 and 84 h of differentiation. In electrophoretic
mobility shift assay, the TSE interacted with activating protein
(AP)-2 in both primary trophoblasts and choriocarcinoma cells. While
CRE-interacting proteins were detectable 12 h after isolation, the
TSE-binding complex did not appear before 36 h of in
vitro differentiation. During syncytium formation increasing
protein expression of activating transcription factor (ATF)-1, cAMP
response element-binding protein (CREB)-1, and AP-2 was observed on
Western blots. Moreover, phosphorylated CREB-1 and ATF-1 accumulated
between 24 and 78 h of trophoblast cultivation. By fluorescence
immunohistochemistry, we show that CREB-1 was predominantly expressed
in syncytiotrophoblasts, whereas ATF-1 and AP-2 localized to the
syncytium and some cytototrophoblasts as well as to stromal and
endothelial cells of the placental villus. Phosphorylated CREB-1/ATF-1
and the coactivator protein CBP were primarily detected in syncytial
nuclei, suggesting the presence of functional, cAMP-dependent
transcriptional complexes in the differentiated tissue. In agreement to
the in vivo situation, phosphorylated CREB-1/ATF-1 were
observed in nuclei of the differentiated trophoblast cultures. The
activity of the CG promoter as well as CREB-1/ATF-1 phosphorylation
increased upon elevation of cAMP levels and overexpression of the
catalytic subunit of protein kinase A. Additionally, we demonstrate
that overproduction of the enzyme enhanced protein expression and
binding of AP-2 to the TSE. We conclude that
differentiation-dependent transcription of the CG gene in villous
trophoblasts is mainly governed by increasing expression of AP-2 and
PKA-dependent phosphorylation of CREB-1 and ATF-1. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-subunit that are common to FSH, LH, and TSH (1). Its
main function is to maintain the corpus luteum during early pregnancy;
however, the biological properties of CG signaled through the CG/LH
receptor in several tissues suggest that the hormone has a more general
role in establishment and maintenance of pregnancy (2, 3, 4, 5).
The major source of CG is the villous trophoblast of the human placenta
(6). Maternal serum levels of CG peak at the ninth to
tenth weeks of gestation and fall to a plateau at low concentrations
thereafter (7). Because concentrations of the free
-subunit increase until parturition, levels of holo-CG are
determined by selective expression of ß-protein (8).
Clinically, elevated serum levels of CG of the second trimester are
known to be associated with chromosomal abnormalities of the fetus such
as Down’s syndrome (9, 10); however, the mechanisms
controlling gestation-dependent expression of CG in normal and
pathological pregnancies remain poorly defined (11).
Because little CG is stored intracellularly, expression/secretion
of the hormone is thought to be mainly triggered by de novo
synthesis of its subunits (12). Numerous studies have
focused on the role of DNA-regulatory elements controlling CG and
CGß messenger RNA (mRNA) transcription in trophoblastic tumor cells
(13). Six genes encoding CGß and a single copy of the
LHß gene are arranged in a gene cluster on chromosome 19 (14, 15). The CGß5 gene is most abundantly transcribed in the
placenta and choriocarcinoma cells (16, 17) and expression
of individual CGß mRNAs varies during the first trimester of
gestation (18). An extended promoter region of the CGß
gene interacting with broadly expressed proteins is required to direct
cAMP-dependent and trophoblast-specific expression
(19, 20, 21). CG mRNA expression in choriocarcinoma cells
is controlled by a composite enhancer harboring five
cis-acting elements located between (-180 and -80)
of the 5' flanking region of the single-copy CG gene,
trophoblast-specific element (TSE), -activating element (ACT),
tandem cAMP response element (CRE), junctional regulatory element
(JRE), and the CCAAT box (Fig. 1A
). The
ACT element is recognized by GATA factors (22) and
comprises the placenta-specific enhancer together with TSE and the CREs
(23, 24). The CREs provide the strongest contribution to
transcription (25, 26) and members of the ATF/CREB
(activating transcription factor/cAMP response element-binding protein)
family were shown to interact with the palindromic sequences in
trophoblastic tumor cells (27). The JRE is located between
the proximal CRE and the CCAAT box, binds an unknown 50-kDa
polypeptide, and augments cell-specific CG expression
(28). The CCAAT box is recognized by -CCAAT binding
factor (CBF), a protein that could be involved in tissue-restricted
expression of a small number of genes (29, 30). Coordinate
expression of CG and ß mRNA may involve activating protein-2
(AP-2), which interacts with the TSE of CG and various sequences of
the CGß promoter (20, 21, 31). However, no
trophoblast-specific DNA-binding protein(s) regulating CG- and
CGß-transcription have been characterized, suggesting that
combinatorial mechanisms may determine tissue-restricted expression of
both subunit mRNAs.
|
mRNA is predominantly found
in the syncytiotrophoblast but can be detected in a small number of
cytotrophoblasts (34). Thus, induction of CG and CGß
mRNA expression in vivo is coupled to differentiation of
cytotrophoblasts to syncytiotrophoblasts. Little information, however,
is available on the mechanisms that control the
differentiation-dependent synthesis of CG subunit mRNAs. In
vitro, CG expression can be studied in mononuclear, villous
cytotrophoblasts isolated from term placenta that aggregate and form
syncytial-like structures in a time-dependent manner
(35, 36, 37). In these cultures, syncytialization is promoted
by a variety of growth factors using cAMP as a second messenger
(38). Indeed, cAMP analogs induced cell fusion in
vitro involving protein kinase A-dependent phosphorylation
(39, 40). During differentiation, an increase in CG
secretion as a consequence of CG and CGß mRNA de novo
synthesis was observed reflecting the in vivo situation
(41, 42). CG rose earlier than CGß during spontaneous
differentiation or upon cAMP stimulation, suggesting that production of
the latter is more strictly dependent on syncytium formation
(43, 44, 45). We have recently demonstrated that the promoter
of the CG gene is active in isolated cytotrophoblasts and increases
during in vitro differentiation (42).
Concomitantly, accumulation of CRE-binding activity harboring
phosphorylated CREB-1/ATF-1 was observed. Here, we extended our studies
on the differentiation-dependent regulation of the CG promoter. In
transient transfection studies, we investigated the contribution of
each DNA-binding site of the CG composite enhancer to
differentiation-dependent transcription and studied the temporal
expression, binding activity and placental localization of
transcription factors involved in CG gene regulation. |
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
wild-type
(wt) or mutant promoters and 0.5 µg pCMV-ß-Gal (CLONTECH Laboratories, Inc., Palo Alto, CA) as a normalization standard.
Two parallel transfections per construct were performed. Precipitates
were left on cells for 12 h and medium was changed. After an
additional 24 h, supernatants were aspirated and cellular protein
lysates were prepared using reporter lysis buffer (Promega Corp.). For analyses of protein kinase A-mediated effects, cells
were transfected with 1 µg plasmid pMtC encoding the catalytic
subunit of protein kinase A (48), 2.5 µg pGL-3 basic
carrying CG wt promoter and 0.5 µg pCMV-ß-Gal after 12 h of
cultivation.
Reporter assays
Luciferase activity and ß-Gal activity were determined in 200
µl of protein extract which had been stored at -70 C in reporter
lysis buffer (Promega Corp.). Luciferase activity was
determined on a luminometer (Lumat LB 9507, EG&G Berthold, Bad
Wildbad, Germany) by using Luciferase Assay System (Promega Corp.). Activity of ß-Gal was quantitated by photometrically
assaying the conversion of the chromogenic substrate chlorophenol
red-ß-D-galactopyranoside (CPRG) (Roche Molecular Biochemicals, Mannheim, Germany) at 570 nm as described
(49). ß-Gal values were determined in 40 µl of protein
lysate incubated 15 min at 37 C. Luciferase- and ß-Gal assays were
performed in duplicate and mean values were calculated. Concentration
of protein lysates was determined by Bio-Rad Laboratories, Inc. Assay Reagent according to the manufacturer’s instructions
(Bio-Rad Laboratories, Inc., Hercules, CA).
Fluorescence immunohistochemistry and immunocytochemistry
For paraffin embedding, cubes of third trimester placental
tissue (15 x 15 x 5 mm) were fixed in a neutral,
phosphate-buffered 4% formaldehyde solution for a maximum of 24 h
at 4 C. The samples were dehydrated in a graded series of ethanol and
embedded in paraffin (melting point 52 C, Merck,
Darmstedt, Germany). Serial sections (3–4 µm) were cut on a rotation
microtome (Microm, Walldorf, Germany) mounted on glass slides and
deparaffinized using xylene and a decreasing series of ethanol (5 min
each step). The slides were heated in microwave oven (2 x 5‘;
850W) in 10 mM citrate buffer (pH 6) for antigen retrieval
pretreatment and the endogenous peroxidase activity was blocked with
3% H2O2 at RT for 15 min.
For immunocytochemistry, purified trophoblasts were plated on chamber
slides and fixed in methanol (-20 C) after 60 h of cultivation.
The staining procedure was performed using Tyramide Signal
Amplification Kit (TSA) for Fluorescence Immunohistochemistry according
to the manufacturer’s instructions (NEN Life Science Products, Boston, MA). Briefly, slides were incubated 30 min in
TNB blocking buffer and exposed to primary antibodies for 1 h at
37 C in a humidity chamber. The following mouse monoclonal antihuman
antibodies were used: cytokeratin 7 IgG1 (clone
OV-TL 12/30, DAKO Corp., Glostrup, Denmark, 0.3 µg/ml),
vimentin IgG2a (clone 3B4, DAKO Corp., 0.5 µg/ml), antidesmosomal protein
IgG1 (Sigma, St. Louis, MO, 18
µg/ml), CG2 IgG1 detecting CG subunit in
CG and free -subunit (Serotec, Oxford, UK, 5 µg/ml),
ATF-1 IgG1 (clone Fl-1, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, 20 µg/ml), CBP
IgG1 (clone C-1, Santa Cruz Biotechnology, Inc., 20 µg/ml). For negative control, the primary antibody
was replaced by isotype IgGs (IgG1,
IgG2a, Serotec, 50 µg/ml). Rabbit
antihuman antibodies (final concentrations) used were: CGß
(DAKO Corp., A0231, 7.6 µg/ml), AP-2 (C-18,
Santa Cruz Biotechnology, Inc., 10 µg/ml), CREB-1 (240,
Santa Cruz Biotechnology, Inc., 10 µg/ml),
antiphosphorylated CREB detecting both phospho-CREB-1 and -ATF-1
(Upstate Biotechnology, Inc., Lake Placid, NY, 1:10), p300
(N-15, Santa Cruz Biotechnology, Inc., 20 µg/ml).
Subsequently, peroxidase-linked antimouse IgG (NA 931, Amersham Pharmacia Biotech, Buckinghamshire, UK, 1:100) or peroxidase
linked antirabbit Ig (NA 934, Amersham Pharmacia Biotech,
1:100), respectively, were added. For costaining of cytokeratin 7 with
p300 or antiphosphorylated CREB, sections were simultaneously incubated
with primary antibodies and, subsequently, with peroxidase linked
antirabbit Ig and antimouse R-phycoerythrin-conjugated goat antibody
(DAKO Corp., 10 µg/ml). After 1 h, slides were
incubated 8 min. in the presence of fluorophore tyramide and mounted
using fluoromount-G (Southern Biotechnology Associates
Inc., Birmingham, AL). To reveal in vitro syncytium
formation, slides that had been incubated with antidesmosomal AB were
counterstained with Hoechst-dye. Finally, all slides were analyzed by
fluorescence microscopy (Olympus Corp., Hamburg,
Germany, BX50) and photographs were taken on Ektachrome 100HC
(Eastman Kodak, Rochester, NY).
Construction of mutant luciferase plasmids
PCR-amplification and subcloning of the CG gene promoter
(-300 to +63) into pGL-3 basic (Promega Corp.) has been
described recently (42). A schematic picture of the 5'
flanking region of the gene carrying the placenta-specific composite
enhancer is shown (Fig. 1A
). Deletion of CREs of the CG promoter was
performed as described (23). Briefly, CG wt promoter
was digested with AatII, cutting palindromic CREs, and
religated. In a second step, constructs harboring a single functional
CRE (CRE-mutant) were digested with AatII, blunted with
mungbean exonuclease and ligated (
CRE-mutant). CCAAT-, JRE-, ACT-
and TSE-pGL-3 constructs were mutated using Transformer Site-directed
Mutagenesis Kit as specified by the supplier (CLONTECH Laboratories, Inc.). Point mutations of TSE, ACT, which
abolish DNA-protein interactions without affecting binding to
neighboring sequences, were described (20, 22). Similarly,
bases that were shown to be critical for binding to CCAAT and JRE
sequences were altered (28, 30). Mutations introduced in
JRE did not affect binding to the proximal CRE in electrophoretic
mobility shift assay (EMSA) experiments (not shown). The following
oligonucleotides were used (mutated bases with respect to the wild-type
are depicted in bold letters, recognition sequences are
marked by brackets): TSE, 5'GCT CCA AAC [AAA AAT
GAT CTG AGA G]TT GAA ACA AGA TAA GAT
C3', ACT, 5'GGG TTG AAA [CAA GAC ATG ATC
A]AA TTG ACG3', JRE, 5'GGT AAA AAT TGA CGT CAT G[AC
GGC CAC A]CC AAG TAC CCT TC3', CCAAT, 5'CCA AGT
ACC CTT GT[A TCT GGC CAT] GGA ATT
TCC TGT TGA TCC3'. All mutants were confirmed by sequencing using the
nonradioactive ABI PRISM Terminator Cycle Sequencing Ready Reaction Kit
(Perkin-Elmer Corp., Foster City, CA). Plasmids were
purified using EndoFree Plasmid Maxi Kit (QIAGEN, Hilden,
Germany).
EMSA
Protein extraction from villous trophoblasts, annealing/labeling
of complementary oligonucleotide sequences, binding reactions, and EMSA
were performed as described elsewhere (42). Extracts were
prepared by freezing (liquid nitrogen)-thawing in buffer containing 20
mM HEPES (pH 7.9), 0.4 M NaCl, 2.5% glycerol,
1 mM EDTA, 1 mM PMSF, 0.5 mM NaF,
0.02 µg/ml leupeptin, 0.02 µg/ml aprotinin, 0.1 µg/ml trypsin
inhibitor, and 0.5 mM DTT. Nuclear extracts from
choriocarcinoma cell lines were performed using Dignam’s method
(50) as described earlier (30). Twenty
micrograms of total protein extract or 5 µg of nuclear extract were
used in binding reactions. Electrophoresis of protein-DNA complexes was
carried out on 5% polyacrylamide gels in the cold (10 V/cm). Gels were
dried and exposed to films (Hyperfilm MP, Amersham Pharmacia Biotech). Oligonucleotides were derived from known DNA-binding
sequences of the CG promoter (20, 51). Sequences of
sense oligonucleotides harboring wild-type (wt) or mutant (mut)
DNA-elements were as follows (mutated bases are depicted in bold
letters, recognition sequences are marked by brackets):
TSE-wt, 5'AC[A AAA ATG ACC TAA GGG] TTG AAA3', TSE-mut, 5'AC[A AAA
ATG ATC TGA GAG] TTG AAA3', CRE-wt,
5'GAT CAA AT[T GAC GTC A]TG GTA A3', CCAAT-wt, 5'AAG TAC CCT TCA
[ATC ATT GGA T]GG AAT TTC CT3', CCAAT-mut, 5'AAG TAC CCT
TGT [ATC TGG CCA T]GG AAT TTC CT3',
AP-2-wt (consensus sequence), 5'GAT CGA ACT GAC CGC CCG CGG CCC GT3'.
In supershift experiments, antibody was added after 30 min of binding
reaction and incubated for an additional 15 min at RT. Antibodies used
were: antiphosphorylated CREB (Upstate Biotechnology, Inc., 1:10), AP-2 (C-18, Santa Cruz Biotechnology, Inc., 100 µg/ml), AP-2ß (C-20, Santa Cruz Biotechnology, Inc., 100 µg/ml), AP-2
(T-17, Santa Cruz Biotechnology, Inc., 100 µg/ml).
Western blot analysis
Cells were cultured for 24, 48, and 72 h, and
proteins were extracted as mentioned above. Equal amounts of protein
(200 µg) were separated on 10% SDS/polyacrylamide gels and
transferred to nitrocellulose (Protran, Schleicher & Schuell, Inc., Dassel, Germany) by electroblotting (Trans-Blot
SD, Bio-Rad Laboratories, Inc.). For
detection of CGß protein 12.5% gels were used. Equal loading of
protein was monitored by Ponceau S staining of membranes. Subsequently,
immunodetection was performed using standard procedures. Final
dilutions of primary and secondary antibodies were as follows:
antiphosphorylated CREB (Upstate Biotechnology, Inc.,
1:1000), 15.2 µg/ml CGß (A0231, DAKO Corp.), 0.1
µg/ml ATF-1/CREB-1 detecting both CREB-1 and ATF-1 (25C10G,
Santa Cruz Biotechnology, Inc.), 0.4 µg/ml AP-2
(C-18, Santa Cruz Biotechnology, Inc.), peroxidase-linked
antimouse IgG (NA 931, Amersham Pharmacia Biotech, 1:100
000), antirabbit Ig horseradish peroxidase (NA 934, Amersham Pharmacia Biotech, 1:50,000). Signals were developed by using
the Enhanced Chemiluminescence System (Amersham Pharmacia Biotech) according to the manufacturer’s instructions. For
reprobing, blots were stripped 10 min in a buffer containing 100
mM 2-mercaptoethanol, 2% SDS and 62.5
mM Tris·Cl, pH 6.7 at 50 C and subsequently
incubated with a different primary antibody. Autoradiographs
were densitometrically scanned and quantification of signals was
done by using Pdi Analysis Software for Biological Data
(Pharmacia Biotech, Uppsala, Sweden). Values obtained
from 24 h cultures were arbitrarily set at 100%. Mean values
(n = 3) at 48 and 72 h were calculated as relative
percentage ± SD.
Statistical analysis
Statistical analyses were performed with Sigma Stat
Statistical Software (Jandel Corp., Chicago, IL) using
Student’s paired t test. A P value <0.05 was
considered statistically significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mRNA were
detectable in immuno-purified trophoblasts after 12 h of
cultivation while CGß mRNA was absent (42, 47). We
concluded that cytotrophoblasts that have recovered from the isolation
procedure are present. Both transcripts were abundantly expressed after
60 h of culturing, indicating functional differentiation.
Therefore, cells were transfected with CG promoter constructs after
12 h and at a later time point (48 h) when numerous syncytia are
detectable in culture (47). Site-directed mutations in the
five DNA elements of the CG composite enhancer were performed to
study their contribution to promoter activity (Fig. 1A). Compared with
the wt promoter, deletion of one CRE did not significantly decrease
luciferase expression in early cultures; however, it reduced reporter
activity to 67% in 48 h transfected cells (Fig. 1B). Mutation of
both CREs diminished luciferase activity to 9% (12 h) and 4% (48 h).
Inactivation of the ACT decreased transcription to 12% and 9%,
while mutation of the JRE reduced reporter expression to 64% and 46%
after 12 and 48 h of transfection, respectively. Mutation of the
CCAAT box decreased promoter activity to 44% in early trophoblast
cultures and to 48% in differentiated cells. After transformation with
the TSE-mutant 81% and 22% of luciferase activity were detected in
extracts prepared after 48 and 84 h of cultivation,
respectively.
In EMSA, the CRE interacts with increasing amounts of phosphorylated
CREB-1/ATF-1, suggesting a role in differentiation-dependent
transcription of the CG promoter (42). JRE, ACT, and
CCAAT contributed significantly to promoter activity in early and late
cultures; however, the above data do not strongly support a role in
differentiation-dependent regulation of the gene. Transfection of the
mutated TSE-reporter at 48 h diminished luciferase expression to a
greater extent than at 12 h, suggesting that the TSE-binding
activity might be required at later stages of in vitro
differentiation. Therefore, we examined DNA-protein interactions of the
TSE in more detail (Fig. 2). TSE was
previously shown to interact with AP-2 from JEG-3 cells
(31). To investigate whether similar complexes can be
found in primary trophoblasts, EMSA using radiolabeled TSE was
performed (Fig. 2A). A low-mobility complex of similar size was
identified in tumor cells and villous trophoblasts. Binding activities
of both cell types were competed by an excess of unlabeled TSE-wt
oligonucleotide, but not in the presence of cold TSE-mutant
oligonucleotide. Interaction with the TSE was abolished upon addition
of an unlabeled AP-2 consensus sequence. The AP-2 complex of JEG-3
cells migrated at the same position as the TSE-complex and could not be
competed by an excess of the mutated TSE oligonucleotide, suggesting
that the two binding activities may contain similar DNA-binding
proteins. Several members of the AP-2 protein family were described, of
which AP-2 was shown to be involved in placenta-specific expression
of the murine adenosine deaminase gene (52). To test which
AP-2 proteins might be present in the TSE-complex, supershift analysis
was performed (Fig. 2B). We found that AP-2 was abundant in the
TSE-complexes of JEG-3 and villous trophoblasts, whereas AP-2ß was
absent. Weak signals were detected in the presence of AP-2
antibodies after long exposure of films (not shown). We failed to
detect AP-2ß and by fluorescence immunohistochemistry, suggesting
that these proteins are either absent or expressed at very low levels
in the human placenta (not shown). Next, we examined the
differentiation-dependent accumulation of the trophoblast-specific
element binding protein (TSEB, Fig. 2C). CRE-binding activity was
readily detectable in 12 h cultures and further increased as
previously described (42). The TSE complex, however, was
absent at 12 h of culturing and steeply increased between 24 and
36 h, suggesting that appearance of the TSE-binding activity is
more closely coupled to the in vitro differentiation
process.
|
). Complexes of
similar sizes were detected in villous trophoblasts, JEG-3 cells, and
JAr cells, which disappeared in the presence of an excess of CCAAT-wt
but not mutant oligonucleotide (Fig. 3, A and B).
|
promoter (Fig. 4A). Both ATF-1 (35
kDa) and CREB-1 (43 kDa) were expressed in early villous trophoblasts.
In three experiments, CREB-1 protein levels increased to 320% ± 90
and 550% ± 210 at 48 and 72 h, respectively, compared with the
amount at 24 h of cultivation (100%). ATF-1 rose to 260% ± 80
(48 h) and 480% ± 180 (72 h). The AP-2 antibody recognized a
single band at 49 kDa in 24 h villous trophoblasts, which
increased during the in vitro differentiation process (440%
± 190 at 48 h and 510% ± 130 at 72 h). Phosphorylated
CREB-1 and ATF-1 appeared in the presence of an antibody reacting with
modified Ser 133 of the kinase inducible domain of CREB-1. The modified
proteins were weakly expressed in cytotrophoblasts. Phosphorylated
CREB-1 and ATF-1, however, increased to 840% ± 350 and 310% ± 140,
respectively, between 48 h (100%) and 72 h of cultivation.
Endogenous CGß protein (27 kDa) was absent in 24 h extracts but
could be detected after 48 h of cultivation (Fig. 4B). Expression
of the ß-subunit further increased at 72 h, indicating
functional differentiation of trophoblasts.
|
). Both CG and
CGß were exclusively detected in the trophoblast epithelium (Fig. 5, C and D). AP-2 protein was strongly expressed in nuclei and
cytoplasm of the syncytium and to a lesser extent in some of the
cytotrophoblasts. Positive signals were also observed in the majority
of stromal and endothelial cells of the villous core (Fig. 5E). ATF-1
localized to nuclei of syncytia, cytotrophoblasts, and endothelia and
was also expressed in some of the mesenchymal cells (Fig. 5F). CREB-1
was found in the cytosol and some nuclei of syncytiotrophoblasts, but
only in a minority of cytotrophoblasts and stromal cells (Fig. 5G).
CREB-1 was undetectable in endothelial cells. Phosphorylated CREB-1 and
ATF-1 were detected in nuclei of the syncytium, but only in few
cytotrophoblasts, underlining the importance of the active
transcription factors in the differentiated villous tissue (Fig. 5H).
Strong expression of the modified proteins was also detectable in
nuclei of stromal cells, whereas nuclei of endothelial cells were
negative. Additionally, we analyzed the adapter proteins p300 and CBP,
which function as coactivators of CREB/ATF-mediated transcription
(53, 54). CBP was strongly expressed in nuclei of
syncytiotrophoblasts and villous stromal cells and to a lesser extent
in nuclei of endothelial cells (Fig. 5I). p300 was primarily expressed
in cytotrophoblasts but could also be detected in a low number of
syncytial nuclei (Fig. 5J). The coactivator was not produced in cells
of the villous core.
|
). At 60 h of differentiation,
CREB-1 was detected in nuclei and cytoplasm, whereas phosphorylated
CREB-1/ATF-1 was preferentially detected in nuclei of cells reflecting
the in vivo situation.
|
promoter activity in early villous trophoblast cultures
(42). Enhanced expression of luciferase was abolished by a
specific inhibitor of protein kinase A (PKA), suggesting that the
activity of the enzyme is required for cAMP-dependent CG mRNA
synthesis. Here, we investigated whether overexpression of the
catalytic subunit of PKA could modulate phosphorylation of CREB-1/ATF-1
and promoter activity of the CG gene (Fig. 7). Cotransfection of a PKA-expressing
plasmid and the CG wt promoter reporter increased luciferase
expression to 430% compared with transfections with CG wt promoter
alone (Fig. 7A). Concomitantly, we observed an increase in
phosphorylated CREB-1 and ATF-1, respectively, on Western blots (Fig. 7B). However, inducibility of promoter was abolished upon
deletion of the tandem CRE (Fig. 7A). Similar results were obtained
upon elevation of cAMP levels by forskolin, an activator of the
adenylate cyclase (Fig. 7, A and B). Interestingly, we also detected an
increase in the protein abundance of AP-2 upon overexpression of PKA
(Fig. 8A). In agreement, the binding
activity of TSE rose upon PKA overexpression (Fig. 8B).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is restricted to a few cell
types, gonadotropes, thyrotropes, and trophoblasts, in accordance
with its coexpression with the different ß-subunit genes of
various glycoprotein hormones (13). Transcription of the
CG gene is controlled by the DNA-elements of the composite enhancer,
which contribute differently to pituitary- and trophoblast-specific
expression (27, 51, 55). CRE and ACT sequences are
required for CG production in JEG-3 and BeWo choriocarcinoma cells
as well as in nontrophoblastic cell lines, whereas the function of JRE
and TSE is restricted to trophoblastic tumor cells (22, 51). Similar to that, the CCAAT element exhibits limited cell
specificity because the particular sequence was shown to be dispensable
for CG transcription in T3–1 gonadotrope cells
(51). Regarding CG mRNA expression in choriocarcinoma
cell lines, extensive studies were performed to characterize the role
of individual DNA elements of the composite enhancer in basal and
cAMP-inducible transcription. Mutations in ACT and JRE sequences
were shown to decrease transcriptional activity to 6.8% and
approximately 42%, respectively, compared with the wt promoter
(22, 51). Deletion of a single CRE reduced transcription
of a CAT-reporter 6-fold (23), whereas removal of both
CREs almost abolished promoter activity (23, 55).
Inactivation of the TSE and CCAAT, respectively, decreased basal
reporter expression in BeWo cells to approximately 63% and 43% of wt
promoter activity (51).
Similar to the data obtained in choriocarcinoma cells, we show that
inactivation of ACT and JRE reduced reporter expression to 12% and
64% in early primary cultures confirming their role in CG
expression of cytotrophoblasts. At a later stage, mutations diminished
transcription to 9% and 46%, respectively, suggesting that the
contribution of the two elements to CG promoter activity does not
largely vary during trophoblast in vitro syncytialization.
Different to choriocarcinoma cells, deletion of one CRE had no impact
on basal CG transcription in early trophoblasts, but reduced
luciferase activity to 67% between 48 and 84 h of cultivation,
suggesting that both copies of the CRE are required for optimal
expression of the gene in differentiated trophoblasts. Inactivation of
the tandem CRE resulted in very low promoter activities (9% and 4%)
underscoring the importance of the tandem element in CG
transcription at different stages of trophoblast differentiation.
Similar to tumor cells, mutation of the CCAAT box reduced transcription
to 44% in early cultures. In late cultures, a reduction to 48% was
observed, suggesting that the element is not involved in
differentiation-dependent CG gene expression. This would also be in
accordance with the data obtained in EMSA because we detected abundant
CCAAT-binding in cytotrophoblasts that remained unchanged during
in vitro differentiation.
Contrary to that, a role of TSE in differentiation-dependent
transcription of the CG gene is suggested, because inactivation of
the recognition sequence contributed little to reporter expression in
early cultures but decreased activity to 22% between 48 and 84 h
of cultivation. Concomitantly, an increase in TSE-binding activity was
observed, which was only detectable at later stages of differentiation,
suggesting that the TSEB fulfills a particular role in cells undergoing
syncytialization. Coordinate expression of CG and CGß genes may
require TSEB because the purified protein was shown to recognize TSE
and different footprint regions of the CGß5 promoter with equal
affinities (20). It has also been shown that TSE is bound
by AP-2, which might be identical to TSEB (20, 31). A role
of AP-2 in coordinating CG expression has been suggested because
overexpression of its mRNA stimulated CG and CGß transcription in
AP-2 deficient cells (31). Here, we demonstrate that the
TSE predominantly interacted with one particular member of the AP-2
family, AP-2, both in primary trophoblast cultures and
choriocarcinoma cells suggesting that the factor could play a major
role in TSE-mediated transcription. Also, protein expression of AP-2
increased during differentiation, which would be compatible with its
requirement in CG expression of differentiated trophoblast
cultures.
CRE-binding activities could be detected in cytotrophoblasts and an
increase between 12 and 36 h of cultivation was noticed preceding
the induction of TSE-complexes. The protein levels of the
cAMP-dependent transcription factors CREB-1 and ATF-1, which we
recently detected in CRE-complexes of the primary cultures
(42), increased during cultivation supporting their role
in differentiation-dependent CG transcription. Phosphorylated,
active forms of CRE-binding proteins were previously identified in EMSA
of early trophoblast extracts using antibodies against phosphorylated
Ser 133 of the kinase inducible domain of CREB-1 (42). The
importance of Ser 133 of CREB-1 in CG transcription has been
described in BeWo cells: compared with the wt, a mutated GAL4-CREB-1
fusion protein was shown to reduce the activity of a promoter in which
the tandem CRE has been replaced by two GAL4 binding sites
(51). Here, we show by Western blot analyses that both
CREB-1 and ATF-1 are phosphorylated in early cultures (24 h),
suggesting that the active forms of both proteins could be involved in
CG transcription of cytotrophoblasts. Upon differentiation, the
phosphorylated forms of CREB-1 and ATF-1 increased supporting the view
that the modified proteins could be necessary for enhanced CG
expression in syncytializing cells. This would also fit to the
observation that the CG promoter required both CREs for full
activity at later stages while a single CRE was dispensable in early
cultures. On tissue sections and in the in vitro cultures
phosphorylated CREB-1/ATF-1 accumulated in nuclei of the syncytium
suggesting the assembly of active transcriptional complexes. Indeed,
the adapter protein CBP was also found in nuclei of
syncytiotrophoblasts, whereas the co-activator p300 mainly localized to
nuclei of residual cytotrophoblasts. From our results, one may conclude
that different cofactors are used by cyto- and syncytiotrophoblasts to
mediate CG expression and transcription of other genes, which
require the intrinsic histone acetyltransferase activity of the
proteins (56). The distinct roles of the two coactivators
in differentiation were previously noticed in other systems (57, 58).
Signaling through cAMP as a second messenger is central to
hormone expression and formation of the syncytium. In primary cultures,
CG and ß mRNA synthesis and secretion as well as transcription
from the CG promoter could be stimulated in response to cAMP
(42, 43, 45). Addition of exogenous cAMP analogs was also
shown to promote in vitro cell fusion of villous
trophoblasts and resulted in down-regulation of the regulatory subunit
of PKA type I (39, 40). Because endogenous cAMP levels
spontaneously rise 6-fold during in vitro syncytium
formation (59), increasing amounts of active forms of the
enzyme might be generated by dissociation/down-regulation of the
inhibitory subunits and translocation of the catalytic protein to the
nucleus, the rate limiting step in transactivation of cAMP-dependent
genes (60). Our data support the role of cAMP/PKA in CG
expression of primary cytotrophoblasts because elevation of cAMP levels
and overexpression of PKA enhanced transcription of the CG wt
reporter as well as steady state levels of phosphorylated CREB-1/ATF-1.
The steeper increase in CREB-1 phosphorylation compared with ATF-1
(Fig. 7B), which has also been noticed in nontransfected,
syncytializing cultures (Fig. 4A), may suggest that phospho-CREB-1
could be the primary mediator of differentiation-/PKA-dependent CG
transcription. Phosphorylated ATF-1, however, may augment or attenuate
the effects of CREB-1. Indeed, in cAMP-responsive cells ATF-1 can also
antagonize CREB-1-dependent transcription by the formation of
heterodimers (61).
Recent data supported the role of AP-2 in mRNA expression of the CGß
gene (31). Three AP-2 sites of the 5' enhancer region
(-318 to -188) are involved in basal transcription since mutation of
the elements reduced promoter activity by 87% (21). The
authors also showed that cAMP-treatment increased binding to the most
proximal AP-2 recognition sequence of the CGß enhancer. Similar to
the CGß promoter, we show that PKA-overexpression augmented binding
to the TSE of the CG composite enhancer and, in correlation to that,
protein expression of AP-2. We assume that PKA-dependent induction
of TSEB-complexes in primary trophoblasts is mainly caused by an
increase in AP-2 transcript levels. Indeed, cAMP was recently shown
to stimulate AP-2 mRNA expression in choriocarcinoma cells
(62).
Phosphorylation of AP-2 by cAMP/PKA-signaling can induce promoter
activity without affecting protein expression and DNA-binding of the
transcriptional activator (63). In tumor cells, however,
it was shown that the AP-2 sites of the CGß enhancer contributed
little to inducible transcription because mutation of the three
recognition sequences only resulted in a 2-fold reduction of
cAMP-stimulation (21). Similar to that, the TSE does not
seem to play a major role in cAMP/PKA-induction of the CG gene in
primary cytotrophoblasts. Because we failed to detect inducible
promoter activity of the tandem CRE-mutant, it is assumed that the CREs
are the primary mediators of cAMP/PKA-responsiveness of early cultures.
In agreement to that, we found that the relative increase of luciferase
activity was similar after coexpression of PKA with CG wt or
TSE-mutant, respectively (not shown).
In conclusion, our data suggest that discrete DNA-elements of the
composite enhancer may contribute differently to basal CG
transcription in cyto- and syncytiotrophoblasts. We suspect that in
differentiating cultures CG promoter activity is mainly governed by
elevation of cAMP and functional PKA which increase CREB-1/ATF-1
phosphorylation and AP-2 expression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Funded by Grant No. 6795 of the "Jubiläumsfond"of
the Austrian National Bank.
Received March 13, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-fetoprotein. Obstet Gynecol 87:217–222[CrossRef][Medline]
- and ß-subunit gene expression in
normal and neoplastic tissues characterized using specific messenger
ribonucleic acid hybridization probes. J Clin Endocrinol Metab 64:319–327- and ß-subunit genes of human chorionic
gonadotropin by trophoblast-specific element-binding protein. Mol
Endocrinol 7:1579–1588-subunit gene in the
placenta and pituitary gland. Mol Cell Biol 14:5592–5602-subunit
gene: dependence on cyclic AMP-inducible elements. Mol Cell Biol 7:3994–4002-subunit gene in the placenta
requires a functional cyclic AMP response element, whereas a different
cis-acting element mediates pituitary-specific expression. Mol Cell
Biol 9:5113–5122-subunit gene occurs through several distinct mechanisms.
Mol Endocrinol 4:573–582-subunit gene is mediated by an 18-base-pair element. Proc
Natl Acad Sci USA 84:2198–2202
subunit gene. J Biol Chem 265:21874–21880 subunit glycoprotein hormone gene utilizes a unique CCAAT binding
factor. J Biol Chem 265:6279–6285Cbf. J Biol Chem 271:21993–22002- and ß-subunit promoters by AP-2. J Biol Chem 272:15405–15412 and placental lactogen mRNAs
during development of the human placenta. J Cell Biol 93:190–198HCG gene promoter in term villous
trophoblasts. Mol Hum Reprod 5:573–580-subunit glycoprotein hormone promoter with trophoblast specificity
and maximal activity. Mol Endocrinol 11:1669–1680
regulates murine adenosine deaminase gene expression during placental
development. J Biol Chem 273:27331–27338This article has been cited by other articles:
 |
|
 |
|
  S. W. Schubert, A. Abendroth, K. Kilian, T. Vogler, B. Mayr, I. Knerr, and S. Hashemolhosseini bZIP-Type transcription factors CREB and OASIS bind and stimulate the promoter of the mammalian transcription factor GCMa/Gcm1 in trophoblast cells Nucleic Acids Res., June 1, 2008; 36(11): 3834 - 3846. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Prast, L. Saleh, H. Husslein, S. Sonderegger, H. Helmer, and M. Knofler Human Chorionic Gonadotropin Stimulates Trophoblast Invasion through Extracellularly Regulated Kinase and AKT Signaling Endocrinology, March 1, 2008; 149(3): 979 - 987. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Sher, N. Yivgi-Ohana, and J. Orly Transcriptional Regulation of the Cholesterol Side Chain Cleavage Cytochrome P450 Gene (CYP11A1) Revisited: Binding of GATA, Cyclic Adenosine 3',5'-Monophosphate Response Element-Binding Protein and Activating Protein (AP)-1 Proteins to a Distal Novel Cluster of cis-Regulatory Elements Potentiates AP-2 and Steroidogenic Factor-1-Dependent Gene Expression in the Rodent Placenta and Ovary Mol. Endocrinol., April 1, 2007; 21(4): 948 - 962. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Knofler, L. Saleh, S. Bauer, B. Galos, H. Rotheneder, P. Husslein, and H. Helmer Transcriptional Regulation of the Human Chorionic Gonadotropin {beta} Gene during Villous Trophoblast Differentiation Endocrinology, April 1, 2004; 145(4): 1685 - 1694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Hirtenlehner, J. Pollheimer, C. Lichtenberger, M. F. Wolschek, H. Zeisler, P. Husslein, and M. Knofler Elevated Serum Concentrations of the Angiogenesis Inhibitor Endostatin in Preeclamptic Women Reproductive Sciences, October 1, 2003; 10(7): 412 - 417. [Abstract] [PDF]
|
|
|
|
 |
|
 S. J. P. Gobin, P. Biesta, J. E. M. de Steenwinkel, G. Datema, and P. J. Van den Elsen HLA-G Transactivation by cAMP-response Element-binding Protein (CREB). AN ALTERNATIVE TRANSACTIVATION PATHWAY TO THE CONSERVED MAJOR HISTOCOMPATIBILITY COMPLEX (MHC) CLASS I REGULATORY ROUTES J. Biol. Chem., October 11, 2002; 277(42): 39525 - 39531. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. M. Thway and M. W. Wolfe An Activator Protein-1 Complex Mediates Epidermal Growth Factor Regulation of Equine Glycoprotein {alpha} Subunit Expression in Trophoblast Cells Biol Reprod, September 1, 2002; 67(3): 972 - 980. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Ben-Zimra, M. Koler, and J. Orly Transcription of Cholesterol Side-Chain Cleavage Cytochrome P450 in the Placenta: Activating Protein-2 Assumes the Role of Steroidogenic Factor-1 by Binding to an Overlapping Promoter Element Mol. Endocrinol., August 1, 2002; 16(8): 1864 - 1880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Peng and A. H. Payne AP-2gamma and the Homeodomain Protein Distal-less 3 Are Required for Placental-specific Expression of the Murine 3beta -Hydroxysteroid Dehydrogenase VI Gene, Hsd3b6 J. Biol. Chem., March 1, 2002; 277(10): 7945 - 7954. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ito, S. Nomura, M. Okada, Y. Katsumata, A. Iwase, F. Kikkawa, M. Tsujimoto, and S. Mizutani Transcriptional regulation of human placental leucine aminopeptidase/oxytocinase gene Mol. Hum. Reprod., September 1, 2001; 7(9): 887 - 894. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Thway and M. W. Wolfe Epidermal Growth Factor Regulation of Equine Glycoprotein Hormone {{alpha}} Subunit Expression in Trophoblast Cells Biol Reprod, July 1, 2001; 65(1): 197 - 203. [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 |
