| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
in the Human Adrenocortical Cancer Cell Line H295R1
Groupe d’Etude en Physiopathologie Endocrinienne, Centre National de la Recherche Scientifique, UPR1524, Institut Cochin de Génétique Moléculaire, Université René Descartes-Paris V, 75014 Paris, France
Address all correspondence and requests for reprints to: Jérôme Bertherat, M.D., Ph.D., Groupe d’Etude en Physiopathologie Endocrinienne, Centre National de la Recherche Scientifique, UPR1524, Institut Cochin de Génétique Moléculaire, Centre Hospitalier Universitaire Cochin, 24 rue du Fg. St. Jacques, 75014 Paris, France. E-mail: bertherat{at}icgm.cochin.inserm.fr
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Western blot and RT-PCR studies demonstrated that CREB is not expressed
in the human adrenocortical cancer cell line H295R, whereas it is
expressed in normal adrenal. During transient transfection
experiments, cAMP stimulation of two reporter genes containing
canonical CRE was maintained. Cotransfection of the dominant negative
inhibitor A-CREB, which prevents transcription factors containing a
CREB-like leucine zipper domain to bind DNA, completely inhibited
cAMP-induced stimulation of CRE activity. Western blot and RT-PCR
studies showed that activating transcription factor-1 (ATF-1),
CRE modulator-
/
(CREM/), and CREM
2 are expressed in
H295R cells. High amounts of CREM proteins were present in H295R,
demonstrating an overexpression of this transcription factor in the
absence of CREB. Furthermore, expression of the activator isoform
CREM was very high in H295R compared to normal adrenal cortex.
Transfection assays demonstrated that CREM2 is a potent
stimulator of CRE activity in H295R. Finally, gel retardation assays
showed that CREM and ATF-1 are the nuclear proteins that specifically
bind the CRE in H295R cells, whereas CREM binding to CRE is not
observed in a CREB-expressing cell line.
H295R cells are the first established nontransgenic cell line that does
not express the ubiquitous transcription factor CREB. H295R
demonstrates that CREM up-regulation can compensate for CREB
deficiency to maintain CRE regulation by cAMP and is a model of
compensation mechanisms between the members of the CREB/CREM/ATF-1
family of transcription factors. This loss of CREB expression and the
overexpression of CREM could be linked to cellular transformation, as
the normal adrenal cortex express high levels of CREB and no or low
levels of CREM.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the adrenal cortex, the pituitary peptide hormone ACTH is the major
activator of the cAMP pathway. ACTH is required for adrenal cortex
activity and steroid synthesis (17). In humans, chronic ACTH
stimulation, as observed, for instance, in Cushing’s disease, leads to
adrenal cortex hyperplasia and cortisol hypersecretion. Furthermore,
activation of the cAMP pathway by mutation of the -subunit of G
proteins (oncogene gsp, which leads to constitutive
activation of adenylyl cyclase) or seven-transmembrane receptor ectopic
expression has been observed in cases of Cushing’s syndrome with
adrenal macronodular hyperplasia or adrenal tumors (18, 19, 20). However,
regulation of the proliferation of adrenocortical cells by ACTH and
cAMP is complex, because an inhibitory role of ACTH has also been
observed in some species (21). Among the various genes that are
stimulated by ACTH and/or cAMP in the adrenal cortex, a classical CRE
has been observed in the promoter of c-fos,
11ß-hydroxylase (CYP11B1), and aldosterone synthase genes (CYP11B2)
(22, 23, 24). By contrast, other cAMP-responsive genes, coding for
steroidogenic enzymes or the ACTH receptor, present cAMP-responsive
regulatory regions that frankly differ from the canonical CRE in their
promoter (25). The H295R cell line was established from an invasive
human adrenocortical carcinoma responsible for Cushing’s syndrome (26, 27). It is considered a model for human adrenal studies, because most
adrenal steroidogenic enzymes are expressed and regulated by cAMP in
this cell line (28).
Our aim was to characterize the CRE-binding proteins in the H295R cell line to assess their roles in the formation of steroid-secreting tumors of the adrenal cortex. Surprisingly, the ubiquitous transcription factor CREB is not expressed in this cell line despite the presence of cAMP-dependent regulation of a classical CRE. ATF-1 and CREM are the major CRE-binding proteins in this cell line, and CREM is up-regulated in the absence of CREB, compensating for CREB deficiency in these adrenocortical tumor cells. H295R is the first cell line in which CREB is not expressed, demonstrating the possibility of compensation within members of the CREB/CREM/ATF-1 family of transcription factors. As CREB is expressed in normal adrenal cortex, where CREM is much lower than in H295R cells, the alterations of this family of transcription factor could take part in adrenal cellular transformation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The human adrenocortical H295R cell line (27, 28) was grown in DMEM-Ham’s F-12 supplemented with 2% Ultroser G (Biosepra), 2 mmol/L glutamine, 5 µg/mL insulin, 5 µg/mL transferrin, 5 ng/mL selenium, 50 U/mL penicillin, and 50 mg/mL streptomycin. The human cell lines DMS 79 and HeLa were grown as previously described (29). H293 cells were cultured in DMEM medium supplemented with 10% FBS, 2 mmol/L glutamine, and 50 mg/L gentamicin. All cells were cultured at 37 C in an atmosphere of humidified air containing 5% CO2. The cell lines were chosen because of their human origin. The HeLa and H293 cells were chosen for their high content of ATF-1 and CREB, respectively. The DMS79 cell line is an ACTH-secreting small cell carcinoma and was chosen as another cause of Cushing’s syndrome. Cells obtained from primary culture of bovine adrenal cortex were a gift from Dr. Jean-Jacques Feige (INSERM U-244, Grenoble, France). Transfections were performed by calcium phosphate coprecipitation in 10-cm dishes. The cells were exposed to the precipitate for 3 h and harvested 48 h later. When appropriate, cells were treated with forskolin (10 µmol/L) and 3-isobutyl-1-methylxanthine (IBMX; 0.5 mmol/L) for 12–15 h. All reagents used for cell culture were purchased from Sigma (St. Louis, MO). Chloramphenicol acetyltransferase (CAT) assays were performed by enzyme-linked immunosorbent assay using a kit from Roche Molecular Biochemicals (Mannheim, Germany). The reporter gene RSV-ßGal (3 µg) was used as an internal control for transfection efficiency, and CAT assays were normalized to ß-galactosidase activity. Transfection experiments were performed in duplicate. Results shown are the means of duplicate determinations for a representative experiment. Each transfection experiment was repeated at least three times with similar results.
Plasmids
The origins of RSV-ßGAL, CG-CAT, somatostatin-CAT, and
c-Fos-CAT reporter plasmids, PKA, Rous sarcoma virus (RSV)-CREB (and
parental PGR expression vectors) (14), the cytomegalovirus
(CMV)-CREM and CMV-CREM2a expression vectors (30), and the A-CREB
and parental pRc/CMV500 expression vectors (31) have been previously
reported. DNA used for transfections was purified using the
QIAGEN-tip 500 kit (QIAGEN, Chatsworth, CA),
according to the manufacturer’s procedures.
Western blots and antibodies
Nontumoral human adrenal cortex were obtained during adrenalectomy performed for adrenal benign nonsecreting adenomas revealed as adrenal incidentaloma. The adjacent nontumoral adrenal cortex was immediately dissected by the pathologist and frozen in liquid nitrogen. The samples were macroscopically devoid of contaminating tissue from the adrenal medulla. Nuclear extracts were performed as previously reported with an addition of okadaic acid (1 µmol/L) in the buffers (14). Protein assays were performed using the Bio-Rad Laboratories, Inc., protein assay kit (Hercules, CA). Proteins were resolved by electrophoresis on 10% SDS-polyacrylamide gel and transferred to nitrocellulose sheets. CREB immunoblots were performed as previously described, using Ab 244 raised against a synthetic peptide spanning residues 128–162, Ab 253 raised against the full-length CREB protein, and Ab 5322 raised against a phosphorylated CREB peptide containing residues 128–141 (32). The CREB 24HB4 specific antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was used for gel shift studies and recognized amino acids 254–327 within the DNA-binding domain of human CREB. The specific ATF-1 C41 and CREM X-12 antibodies were purchased from Santa Cruz Biotechnology, Inc.. For Western blot detection of the first antibody, the Immun-Star goat antirabbit IgG detection kit (Bio-Rad Laboratories, Inc.) was used to detect CREB and CREM antibodies. To detect the ATF-1 C41 antibody, an antimouse IgG-horseradish peroxidase and the enhanced chemiluminescence kit (Amersham Pharmacia Biotech, Arlington Heights, IL) were used. Bacterial recombinant CREB was a gift from Dr. M. Montminy; recombinant ATF-1 (amino acids 39–271) and recombinant full-length CREM were purchased from Santa Cruz Biotechnology, Inc..
RNA extraction, RT-PCR, cloning, and sequencing
Polyadenylated messenger RNA (mRNA) was extracted from H295R and
DMS79 using the Micro FastTrack kit (Invitrogen, San
Diego, CA). RT was performed with the complementary DNA (cDNA) cycle
kit (Invitrogen). cDNA was amplified by PCR using the
Dynazyme II DNA polymerase (Finnzymes Oy, Finland) for 35 cycles
(94 C for 1 min, 56–60 C for 2 min, and 72 C for 3 min). As a negative
control for the RT-PCR reaction, reverse transcriptase was omitted from
the reaction mixture. The PCR products were separated on an agarose gel
and stained with ethidium bromide. For Southern analysis, PCR products
were transferred from a 1.5% agarose gel by capillary blotting to a
nylon membrane (Hybond-N+, Amersham Pharmacia Biotech). Blots were hybridized at 5 C below the
oligo(Tm), overnight with specific oligonucleotides end labeled with T4
polynucleotide kinase and [-32P]ATP.
Membranes were washed at room temperature twice for 10 min each time
with 2 x SSC (standard saline citrate)-0.5% SDS for 30 min at
hybridization temperature with 2 x SSC-0.5% SDS and for 20 min
at hybridization temperature with 1 x SSC-0.5% SDS.
The synthetic oligonucleotides (Life Technologies, Inc.,
Gaithersburg, MD) used were as follows. For CREB cDNA amplification,
the letter denotes the exon named according to Hoefler et
al. (33) (Fig. 2
): sense primers: B, 5'-ACC ATG GAA TCT GGA GCC
GAG AAC-3'; G, 5'-TAC CCA GGG AGG AGC AAT ACA GCT-3'; and antisense
primers: I, 5'-GTG GCA GTA AAG GTC CTT AAG TGC-3'; E, 5'-CTG TAG GAA
GGC CTC CTT GAA AGA-3'; G, 5'-TTG AAC AAC AAC TTG GTT GCT GGG-3'. The
hybridization oligonucleotides used and labeled as probe for CREB
Southern blotting were as follows: E, 5'-TTT CAA CTA TTG CAG AAA GTG
AAG-3'; F, 5'-TAT ACT GTC CAC TGC TAG TTT GGT-3'; and G, 5'-ATG GTT AAT
GTT TGC AGG CCC TGT-3'. The cDNA amplification primers used for actin
have been reported previously (34). The primers used for ATF-1 cDNA
amplification were: sense, 5'-GAA GAT TCC CAC AAG AGT ACC ACG-3'; and
antisense, 5'-CAG GAC TGC AAC TCG GTT TTC CAG-3'. The primers used for
CREM cDNA amplification were: sense, 5'-GAC CAT GGA AAC AGT TGA ATC
CCA-3' [located in exon B, according to Gellersen et al.
(30)]; and antisense, 5'-AAA CTT CCG GGC GAT GCA GCC ATC-3' (located
in exon H) and 5'-CGT CGA CAT TCT TTG GCA GC-3' (located in exon
J).
|
|
Electrophoretic mobility shift assays
Gel shift assays with nuclear extracts from H295R and DMS79 were
incubated in binding buffer [20 mmol/L HEPES, 0.2 mmol/L
ethylenediamine tetraacetate, 0.1 mmol/L NaCl, 0.1 mmol/L KCl, 10
mmol/L MgCl2, 200 µg/mL BSA, 35% glycerol, and
5 µg poly(dI-dC)]. Labeled probe (5 fmol) was added to the reaction
buffer, which contained 5 µg nuclear extracts, and was incubated at
room temperature for 30 min. The complexes were resolved on a 6%
nondenaturing polyacrylamide gel using an electrophoresis TBE 0.5x
buffer. The gel was then dried, and the protein-DNA complexes
were visualized autoradiographically using BIOMAX film (Eastman Kodak Co., Rochester, NY). For antibody supershift assays the
nuclear extracts, antibody (1 µL 24HB4, C41, or X12) and binding
buffer were preincubated for 30 min at room temperature before a 30-min
incubation with labeled probe. The synthetic annealed oligonucleotides
were end labeled with T4 polynucleotide kinase and
[-32P]ATP. The sequences of the wild-type
and mutant oligonucleotides were as follows: rat somatostatin CRE,
5'-CGC CTC CTT GGC TGA CGT CAG AGA-3' (1); mutant somatostatin, 5'-CGC
CTC CTT GGC TAA CAT CAG AGA-3'; human CG CRE, 5'-GAT CCG GCT GAC GTC
ATC AAG CTA-3' (35); NFY, 5'-GGG GTA GGA ACC AAT GAA ATG AAA GGT
TA-3' (36); human collagenase activating protein-1 (AP-1),
5'-GAT CAA AGC ATG CAG ACA CCT-3' (37).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To study CREB expression and CREB phosphorylation in H295R,
Western blot and RT-PCR studies were performed. Western blot with two
different CREB antibodies (244 and 253) showed no signal using nuclear
extracts of H295R, demonstrating clearly the lack of CREB protein in
this cell line (Fig. 1
, A and B). In
contrast, CREB protein was easily detected in control cells (H293,
DMS79, and HeLa) as well as in nontumoral human and bovine adrenal
cortex (Fig. 1, C and D). Western blot with the 5322 phospho-CREB
antibody showed that CREB is highly phosphorylated in H293 and DMS79,
but that low levels of phosphorylation of CREB are present in
unstimulated HeLa cells. The lack of signal with the nonphosphorylated
recombinant CREB demonstrates the specificity of the 5322 antibody for
the phosphorylated form of CREB (Fig. 1B). However, a lower molecular
weight protein was detected with the 5322 antibody in H295R, HeLa, and
H293 cells. This protein is probably phospho-ATF-1. CREB and ATF-1
share high sequence homology in their domains surrounding the PKA
phosphorylation site (Ser133 in the KID
domain of CREB, and Ser63 in ATF-1). Thus, the
5322 antibody should recognize phospho-ATF-1 as well as phospho-CREB
(38).
To determine whether the lack of CREB protein was due to the loss of
CREB gene expression in H295R, RT-PCR studies were
performed. Using cDNA transcribed from H295R mRNA we found no PCR
signal with three different sets of primers that amplified most of the
CREB cDNA or its 5'- or 3'-regions. In contrast, all three sets of
primers detected the expected CREB mRNA signal in the control cell line
DMS79 (Fig. 2).
Despite the lack of CREB, transcriptional regulation by cAMP is preserved in H295R
The lack of CREB expression might lead to the loss of
cAMP-dependent transcription. Thus, the activities of two different
promoters, both containing canonical CREs, were studied in H295R.
Cotransfection of the catalytic subunit of PKA or treatment with
forskolin and IBMX to increase cAMP levels both stimulated the activity
of reporter genes containing the somatostatin or the CG promoters
(Fig. 3A). The c-Fos promoter was also
responsive to cAMP in this cell line (data not shown). The stimulation
of the somatostatin promoter by forskolin and IBMX was similar in H295R
and H293 cells, which express CREB (data not shown). This demonstrates
that the cAMP transcriptional response was maintained despite the lack
of CREB in the H295R cell line.
|
CG promoters (Fig. 3B). As A-CREB specifically
binds to members of the CREB/ATF-1 family and not to other
transcription factors from the b-ZIP family (31), this response
demonstrates that ATF-1 and/or CREM compensate for the lack of CREB in
H295R. ATF-1 and several isoforms of CREM are expressed at high levels in H295R
To collect further evidence that CREM and/or ATF-1 can compensate for CREB deficiency in H295R, CREM and ATF-1 gene expression and protein levels were studied in this cell line and compared to those in other established human cell lines.
Western blot using an ATF-1-specific antibody showed a single band,
demonstrating that ATF-1 protein is present in H295R (Fig. 4B). The amount of ATF-1 protein in H295R was
almost as high as that in HeLa cells. This observation is in agreement
with the phospho-ATF-1 protein detected using the 5322 phospho-CREB
antibody (Fig. 1B). The HeLa cells contained the highest level of ATF-1
protein among all cell lines that we tested (data not shown). RT-PCR
performed with cDNA transcribed from H295R and DMS79 mRNA, using a set
of primers amplifying most of the coding part of ATF-1, showed a single
band of the expected size in both cell lines (Fig. 4A).
|
). CREM
protein was not detectable in DMS 79, a low amount of CREM protein was
observed in HeLa cells, and a very high amount of CREM protein was seen
in H295R cells. The respective molecular weights indicate that the two
major forms of CREM protein detected in H295R correspond to human
CREM2 (higher molecular weight protein) and CREM or -
(lower molecular weight proteins; Fig. 5A). By contrast, in normal
human adrenal cortex (Fig. 5B), the level of CREM protein was much
lower than the level of CREM/.
|
, C and D). Sequencing of
cloned PCR products amplified with primers B-H identified six isoforms
of CREM. Four of these isoforms (CREM
1,2; CREM1P, CREM1,P2;
and CREME) contain a premature stop codon (30). Except for CREM
E, all of these isoforms have previously been observed in other cell
lines. The CREM E isoform lacks exon E and is predicted to encode a
truncated protein missing the DNA-binding domain of CREM. Direct
sequencing of PCR products amplified with primers B–J identified the
remaining isoforms as CREM2 and CREM/. These isoforms do
not contain a premature stop codon and correspond to the proteins
detected by Western blot. The mRNA of CREM contains the two
DNA-binding domains of CREM (DBDI and DBDII). As a stop codon is
present at the 3'-end of DBDI (30), two proteins of similar molecular
weight were expected for the CREM and CREM isoforms. Major differences are observed between the nuclear CRE-binding proteins in a cell line expressing CREB and H295R cells
To specify the pattern of CRE-binding proteins in H295R, electrophoretic mobility shift assays were performed. Electrophoretic mobility shift patterns were compared between the CREB-deficient H295R cell line and the neuroendocrine DMS 79 cell line expressing CREB.
Gel shift experiments were performed using probes corresponding to the
CRE sequence of the somatostatin and CG promoters. The pattern of
the complexes was dramatically different in the two cell lines. Three
specific complexes were identified in both cell lines, as demonstrated
by competition with the nonspecific NFY probe (Fig. 6, labeled 1, 2, and 3 in H295R and 4, 5, and
6 in DMS 79). These complexes were abolished by the addition of
nonradiolabeled probes containing a canonical CRE (CG or
somatostatin CREs) or a closely related CRE (c-Fos CRE). Using H295R
nuclear extracts, complex 3 was partially inhibited after addition of
unlabeled probes containing a mutated CRE (mutSOM) or the AP-1
oligonucleotide, suggesting that part of the nuclear proteins
comigrating with complex 3 are specific canonic CRE-binding proteins in
this cell line. In contrast, using DMS 79 nuclear extracts, complex 6
was completely abolished after addition of unlabeled probes containing
a mutated CRE (mutSOM) or an AP-1 oligonucleotide, demonstrating that
an intact canonic CRE is not required for formation of this
complex.
|
, A, right, and B) were performed
to identify the proteins responsible for each specific complex. In
H295R, complex 1 was abolished by a specific ATF-1 antibody. Complexes
2 and 3 were abolished by a CREM antibody. For complex 3, which is
partly made up of noncanonical CRE-specific binding, the shift was more
evident after the addition of an excess mutant somatostatin CRE (Fig. 6B). No supershift was observed in H295R using the specific CREB
antibody, demonstrating that CREB is not part of the CRE-binding
proteins in this cell line, as expected. In contrast, complex 4 in DMS
79 was abolished by the CREB antibody. These results confirm that ATF-1
and CREM are the major proteins that bind the CRE in H295R, as expected
from the transient transfection assays using the dominant negative
mutant A-CREB. Regulation of cAMP-dependant transcription by CRE-binding proteins in the absence of CREB
Several antagonist trans-activation potentials have
been described for the various CREM isoforms, and CREB deficiency could
modify these potentials. We, therefore, studied CRE regulation by CREM
in H295R. Cotransfection of the somatostatin reporter gene and CREM
or CREM2 expression vectors was performed in this cell line (Fig. 7). CREM slightly inhibited the
forskolin-induced stimulation of CRE activity. This CREM isoform is
usually considered a negative regulator of cAMP-dependent transcription
because it lacks the glutamine-rich domains Q1 and Q2. However, the
magnitude of this inhibition was much less than that observed with the
dominant negative mutant A-CREB. Transfection of a CREB expression
vector increased cAMP stimulation of CRE activity. This increase was
abolished by cotransfection of CREM. Taken together, these data
suggest that CREM is a more potent CRE inhibitor in the presence of
CREB, as expected for a competitive inhibitor. On the other hand,
CREM2 was as effective as CREB in stimulating cAMP-dependent
transcription, as expected, because the CREM2 isoform contains a
glutamine-rich Q2 domain.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because of the central role of ACTH- and cAMP-dependent transcription in adrenal cortex differentiation and proliferation and the potential role of the cAMP pathway in adrenal hyperplasia and tumorigenesis (18, 19, 20), we attempted to characterize the CRE-binding proteins in the H295R cell line. Surprisingly, the ubiquitous transcription factor CREB was not present in this cell line despite its expression in normal human and bovine adrenal cortex. This lack of CREB could be explained by the loss of CREB gene expression, as demonstrated by RT-PCR. Considering the various antibodies used for the Western blot studies and the different sets of PCR primers used, it seems unlikely that a truncated form of CREB is expressed in this cell line, as previously reported in other tissues (41). H295R is the first cell line identified to date that does not express the CREB gene. The F9 embryonal carcinoma cell line has a defect in cAMP-induced CRE activity that is compensated for by overexpression of exogenous CREB and PKA, suggesting that CREB is mandatory for cAMP-dependent transcription (5).
CREB is ubiquitously expressed (reviewed in Ref. 3) and is thought to
be necessary for cAMP-dependent transcription through the CRE.
Microinjection of CREB antibody abolishes CRE activity in response to
PKA stimulation (7). The H295R cell line is the first nontransgenic
model in which cAMP or PKA stimulation of CRE activity can occur in the
total absence of CREB. We have demonstrated that the closely related
transcription factors CREM and ATF-1 bind the CRE in this
adrenocortical cell line and compensate for the loss of CREB
expression. The complete inhibition of CRE activity by the dominant
negative mutant A-CREB supports this conclusion. The acidic extension
of A-CREB interacts specifically with the basic region of transcription
factors containing a CREB-like leucine zipper (31). A-CREB then blocks
binding of CREB and closely related transcription factors such as ATF-1
to DNA. In contrast, A-CREB does not modify binding of the b-ZIP
factors of the AP-1 complex to DNA. ATF-1 is considered to be a
positive regulator of cAMP-dependent transcription, whereas CREM
isoforms could be either negative or positive regulators of CRE
activity. We have verified in H295R cells that CREM2 is as
effective as CREB in stimulating CRE activity. However, CREM, which
is a strong inhibitor in the presence of CREB, has only a weak
inhibitory effect in H295R cells. Together the mobility shift assays
and the transfection assays performed in H295R cells demonstrate that
CREM and ATF-1 are the transcription factors compensating for CREB
deficiency in this cell line.
Western blot showed that the levels of CREM proteins are much higher in H295R than in any other cell line that we studied. ATF-1 protein levels were similar in H295R cells and HeLa cells. HeLa cells contain the highest amount of ATF-1 protein among the human cell lines that we studied (42). Therefore, up-regulation of CREM and to a lesser extent of ATF-1 in H295R cells could compensate for CREB deficiency. A similar increase in CREM expression has been observed in CREB null mice (4, 43, 44). CRE activity in these mice has not been reported, but it is tempting to speculate that up-regulation of CREM could compensate for CREB gene deletion. The mechanism leading to CREM up-regulation in the H295R cells remains to be determined. However, in view of the similar expression pattern of the CRE-binding proteins in the CREB null mice and the H295R cells, an inhibition of CREM by CREB could be postulated.
Considering the major role of ACTH and cAMP in adrenal cortex
differentiation and proliferation (16, 17), the dramatic changes in the
expression of the CRE-binding protein of the CREB/ATF-1/CREM family in
H295R cells raises questions about the involvement of these
transcription factors in adrenal tumorigenesis. The adrenal pathology
of the CREB knockout mice has not been reported (4), but the perinatal
lethality observed in these animals might impair such study. In the
normal adrenal cortex CREB protein can be detected, and the level of
the activator CREM is much lower than the level of CREM/
compared to the high CREM expression observed in H295R. In addition
to the role played by cAMP in the proliferation of various endocrine
tissues, a role for CRE-binding proteins in cellular proliferation and
regulation of the cell cycle has also been postulated (12, 13, 14). The
characterization of CRE-binding protein expression in adrenocortical
tumors would be very interesting to study their role in adrenal tumor
formation. In the adrenal cortex, ACTH stimulates the expression of the
protooncogene c-fos (23). The transcriptional stimulation of
c-fos by cAMP involves a CRE (22). Furthermore, most of the
adrenal steroidogenic enzymes are regulated at the transcriptional
level by cAMP. This regulation occurs through canonical CRE or some
cAMP-responsive sequences that diverge from the palindromic CRE (24, 40). Our study demonstrates that the classical CRE is regulated by cAMP
in the same manner in CREB-deficient adrenal H295R cells than in a cell
line expressing CREB. Nevertheless, the alterations in the
CREB/ATF-1/CREM family of transcription factors in H295 could have
varying effects on noncanonical CREs.
In summary, the adrenocortical H295R cell line is the first established
cell line that has lost the expression of the ubiquitous transcription
factor CREB. It therefore represents a unique model for studying the
regulation of transcription by cAMP in the absence of CREB. The
isoforms CREM and CREM of the related transcription factor
CREM are highly expressed in this cell line. ATF-1 and CREM are the
nuclear proteins that bind to the CRE in H295R and compensate for the
lack of CREB expression to maintain CRE regulation by cAMP. This human
adrenocortical cell line demonstrates that CREB is not mandatory for
CRE regulation by cAMP, because of the complex interplay between
members of the CREB/CREM/ATF-1 family of transcription factors.
Finally, as normal adrenal cortex and various human cell lines that we
have tested express CREB at a high level and CREM at a low level
compared to H295R, an alterations of the transcription factors of the
CREB/CREM/ATF-1 family could take part in adrenal cortex tumor
formation.
|
Acknowledgments |
|---|
and 2) plasmids, Drs. W. Rainey and C.
Gicquel for providing the H295R cell line, Dr. A. Louvel for dissection
of the human adrenal tissue, Dr. M. A. Dugue and the staff of the
Laudat Laboratory for steroid assays of H295R culture media, F.
Letourneur and E. Gomas for sequencing, and Drs. M. G. Catelli, P.
Brindle and members of our laboratory for invaluable discussions. |
Footnotes |
|---|
Received June 28, 1999.
Revised August 30, 1999.
Accepted September 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene transcription is directed
by a repeated 18-base pair enhancer. -promoter receptivity to the
enhancer confers cell-preferential expression. J Biol Chem. 262:12169–12174.This article has been cited by other articles:
 |
|
 |
|
  Testosterone patches for female sexual dysfunction DTB, March 1, 2009; 47(3): 30 - 34. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Rizk-Rabin, G. Assie, F. Rene-Corail, K. Perlemoine, H. Hamzaoui, F. Tissier, M. Lieberherr, X. Bertagna, J. Bertherat, and Z. Bouizar Differential Expression of Parathyroid Hormone-Related Protein in Adrenocortical Tumors: Autocrine/Paracrine Effects on the Growth and Signaling Pathways in H295R Cells Cancer Epidemiol. Biomarkers Prev., September 1, 2008; 17(9): 2275 - 2285. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O Gubbay, M T Rae, A S McNeilly, F X Donadeu, A J Zeleznik, and S G Hillier cAMP response element-binding (CREB) signalling and ovarian surface epithelial cell survival. J. Endocrinol., October 1, 2006; 191(1): 275 - 285. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Sugawara, N. Sakuragi, and H. Minakami CREM confers cAMP responsiveness in human steroidogenic acute regulatory protein expression in NCI-H295R cells rather than SF-1/Ad4BP. J. Endocrinol., October 1, 2006; 191(1): 327 - 337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Libe and J. Bertherat Molecular genetics of adrenocortical tumours, from familial to sporadic diseases Eur. J. Endocrinol., October 1, 2005; 153(4): 477 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Tissier, C. Cavard, L. Groussin, K. Perlemoine, G. Fumey, A.-M. Hagnere, F. Rene-Corail, E. Jullian, C. Gicquel, X. Bertagna, et al. Mutations of {beta}-Catenin in Adrenocortical Tumors: Activation of the Wnt Signaling Pathway Is a Frequent Event in both Benign and Malignant Adrenocortical Tumors Cancer Res., September 1, 2005; 65(17): 7622 - 7627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. de Fraipont, M. El Atifi, N. Cherradi, G. Le Moigne, G. Defaye, R. Houlgatte, J. Bertherat, X. Bertagna, P.-F. Plouin, E. Baudin, et al. Gene Expression Profiling of Human Adrenocortical Tumors Using Complementary Deoxyribonucleic Acid Microarrays Identifies Several Candidate Genes as Markers of Malignancy J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1819 - 1829. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J Liu, X-D Li, A Ora, P Heikkila, A Vaheri, and R Voutilainen cAMP-dependent protein kinase activation inhibits proliferation and enhances apoptotic effect of tumor necrosis factor-{alpha} in NCI-H295R adrenocortical cells J. Mol. Endocrinol., October 1, 2004; 33(2): 511 - 522. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bertherat, L. Groussin, F. Sandrini, L. Matyakhina, T. Bei, S. Stergiopoulos, T. Papageorgiou, I. Bourdeau, L. S. Kirschner, C. Vincent-Dejean, et al. Molecular and Functional Analysis of PRKAR1A and its Locus (17q22-24) in Sporadic Adrenocortical Tumors: 17q Losses, Somatic Mutations, and Protein Kinase A Expression and Activity Cancer Res., September 1, 2003; 63(17): 5308 - 5319. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Gu, Y. Wen, A. Mison, and J. L. Nadler 12-Lipoxygenase Pathway Increases Aldosterone Production, 3',5'-Cyclic Adenosine Monophosphate Response Element-Binding Protein Phosphorylation, and p38 Mitogen-Activated Protein Kinase Activation in H295R Human Adrenocortical Cells Endocrinology, February 1, 2003; 144(2): 534 - 543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-F. Louet, G. Hayhurst, F. J. Gonzalez, J. Girard, and J.-F. Decaux The Coactivator PGC-1 Is Involved in the Regulation of the Liver Carnitine Palmitoyltransferase I Gene Expression by cAMP in Combination with HNF4alpha and cAMP-response Element-binding Protein (CREB) J. Biol. Chem., October 4, 2002; 277(41): 37991 - 38000. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Doi, H. Takemori, X.-z. Lin, N. Horike, Y. Katoh, and M. Okamoto Salt-inducible Kinase Represses cAMP-dependent Protein Kinase-mediated Activation of Human Cholesterol Side Chain Cleavage Cytochrome P450 Promoter through the CREB Basic Leucine Zipper Domain J. Biol. Chem., May 3, 2002; 277(18): 15629 - 15637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bailey, R. J. Phillips, A. J. Pollard, K. Gilmore, S. C. Robson, and G. N. Europe-Finner Characterization and Functional Analysis of cAMP Response Element Modulator Protein and Activating Transcription Factor 2 (ATF2) Isoforms in the Human Myometrium during Pregnancy and Labor: Identification of a Novel ATF2 Species with Potent Transactivation Properties J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1717 - 1728. [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 |
