Download as a PDF

Two Separate Mechanisms for
Ligand-Independent Activation of
the Estrogen Receptor
Mohammed K. K. El-Tanani and Chris D. Green
School of Biological Sciences
University of Liverpool
Liverpool, L69 3BX, United Kingdom
mone activates the receptor protein so that it may bind
to discrete estrogen response element (ERE) sequences in the genomic DNA and stimulate the transcription of specific structural genes. The ER, in common with other members of the nuclear receptor
superfamily, possesses a modular structure in which
various aspects of receptor function are associated
with specific regions of the peptide sequence (4, 5). In
addition to well defined DNA-binding and ligand-binding domains, nuclear receptors possess two separate
regions that are required for optimal transcriptional
activation. An amino-terminal transcription activation
function (AF-1) operates in a manner that is independent of ligand binding (6, 7). A second activation function (AF-2) is located in the ligand-binding domain of
the receptor toward the carboxy terminus of the molecule, and its activity is dependent upon the binding of
an agonistic ligand (6, 7). Both AF-1 and AF-2 are
required for optimal stimulation of transcription, but
the relative contributions of the two varies in a promoter- and cell type-specific manner (6–8).
However, more recent evidence has suggested that
the ER, in particular, may be activated in a ligandindependent manner by a variety of agents that include several growth factors (9–11), the neurotransmitter dopamine (12), and cAMP (13, 14). The
expression of a variety of estrogen-responsive genes
has been shown to be stimulated by these agents by
a mechanism that is inhibited by the presence of the
pure antiestrogen, ICI 164384 (9, 11, 13). Because ICI
164384 acts by binding to the ER (15), this is taken as
evidence that transcriptional activation by these alternative agents involves the ER. Stimulation by epidermal growth factor (EGF) of the expression of estrogenresponsive reporter gene constructs in HeLa cells,
which do not express endogenous ER, has been
shown to be dependent upon coexpression of ER (9).
In addition, it has been shown that the estrogen-like
effects of EGF upon the mouse uterus do not occur in
ER-deficient transgenic mice (10).
We now present evidence that the activation of the
human ER by EGF and by cAMP involves mechanisms
distinct from that of estradiol and distinguishable from
each other. Furthermore, the transactivation effect of
Transient transfection experiments in which three
different estrogen response element-containing
reporter genes were cotransfected into HeLa cells,
together with constitutively expressed estrogen
receptor (ER) constructs, demonstrate that activation of the transcription of the reporter genes by
epidermal growth factor (EGF) and by cholera toxin
with 3-isobutyl-1-methyl-xanthine, which elevate
cellular cAMP, is dependent upon the presence of
functional ER. Cotransfection of the reporter genes
with truncated versions of the ER shows that the
two non-ligand activators of ER require different
regions of the receptor to produce their effects on
transcription. EGF acts primarily by means of
transactivation domain AF-1, whereas cAMP acts
via transactivation domain AF-2 of the ER. A point
mutation that removes a major site of inducible
phosphorylation within the AF-1 domain of the ER
abolishes the response to EGF, but the response to
estradiol and cAMP is retained. Specific inhibition
of cAMP-activated protein kinase (protein kinase
A) prevents the response to elevated cAMP but
does not affect EGF or estradiol responses. Overexpression of the protein kinase A catalytic subunit
in HeLa cells results in an amplified response to
estradiol, similar to that induced by cholera toxin
with 3-isobutyl-1-methyl-xanthine. Comparable
experiments performed using COS-1 cells produce
similar results but also reveal cell type- and promoter-specific aspects of the activation mechanisms. Apparently, the ER may be activated by
three different signal molecules, estradiol, EGF,
and cAMP, each using a mechanism that is distinguishable from that of the others. (Molecular Endocrinology 11: 928–937, 1997)
INTRODUCTION
The estrogen receptor (ER) is a member of a superfamily of nuclear receptors that function as ligandactivated transcription factors (1–3). Binding the hor0888-8809/97/$3.00/0
Molecular Endocrinology
Copyright © 1997 by The Endocrine Society
928
Ligand-Independent Activation of ER
929
EGF is not dependent upon estradiol effects upon
receptor activity.
RESULTS
HeLa cells from human cervical carcinoma do not
contain ERs, but their growth is stimulated by EGF
(16). We transfected eukaryotic expression vectors
containing cDNAs derived from the human ER (4) into
HeLa cells. HEG0 contains an insert coding for the
full-length wild type human ER (amino acids 1–595)
(17). HE0 codes for a full-length ER that contains a
single point mutation resulting in an amino acid substitution, Gly400, which is replaced by a Val residue
(17). HE457 codes for a full-length ER containing a
point mutation resulting in the replacement of Ser118
by an Ala residue (18). HE15 contains an insert coding
for a carboxyl-terminal truncated receptor (amino acids 1–282). HEG19 codes for an amino-terminal truncated receptor (amino acids 179–595). As indicated in
Fig. 1a, the product of HE15 consists of the A/B and C
domains and therefore retains AF-1, but not AF-2,
transactivation function. The HEG19 product consists
of domains C, D, and E/F and therefore has ligandbinding activity and retains AF-2 but not AF-1 transactivation function. We investigated the ability of these
ER constructs to stimulate the expression of chloramphenicol acetyltransferase (CAT) reporter gene constructs, in response to estradiol, EGF, and elevated
cAMP. The relative contributions of AF-1 and AF-2 to
overall transcriptional activation are reported to be
promoter specific (6–8), and we therefore used three
reporter constructs, each with a different promoter
structure (Fig. 1b).
In the absence of ER, the expression, in HeLa cells,
of an estrogen-responsive reporter gene is not stimulated by estradiol, EGF, or cholera toxin 1 3-isobutyl1-methylxanthine (CT/IBMX) (Fig. 2). CT causes irreversible stimulation of adenylate cyclase, and IBMX
inhibits cAMP phosphodiesterase; together they
cause a major increase in intracellular cAMP (13). Cotransfection of the full length ER (HEG0) with the reporter gene enables all three agents to stimulate CAT
expression (Fig. 2). EGF alone causes an increase in
CAT expression that is less than that seen with estradiol; however, simultaneous treatment of the cells with
EGF and estradiol results in a response that is the sum
of the two separate responses. CT/IBMX alone causes
only a weak stimulation of reporter gene activity but
has a synergistic effect when combined with estradiol.
The Gly-.Val mutation in HE0 has been reported to
prevent the ligand-independent activation of the ER by
dopamine (12). However, we found that HE0 is able to
support a pattern of response to estradiol, EGF, and
CT/IBMX, alone and in combination, similar to that
seen in the presence of the wild type receptor. The
pattern of response to the inducers is similar for the
different reporter gene constructs, although the scale
of the increase varies.
Fig. 1. DNA Constructs Transfected into HeLa and COS-1
Cells
a, ER derivatives. Amino acid positions are numbered from
the amino terminus of the ER and indicate the boundaries of
the functional domains of the receptor and of the truncated
receptors. Positions of point mutations (amino acid substitutions) in HE0 and HE457 are indicated. b, Reporter constructs
(not drawn to scale). Solid squares indicate consensus EREs.
ERE.VIT contains two contiguous (nonconsensus) endogenous EREs together with an additional consensus ERE inserted upstream of the endogenous EREs (31). VA, An NF1related vitellogenin activator element (33). 2ERE.TATA is a
synthetic enhancer/promotor sequence containing the elements indicated (31). ERE.TK consists of a consensus ERE
linked to the promotor sequence (nucleotides 2150 to 151)
of the herpes simplex virus thymidine kinase gene (6).
Cotransfection of the truncated receptors, HE15
and HEG19, with the reporter genes reveals differences between the three inducers and their relationships with each other. EGF is able to induce an increase in CAT expression in the presence of HE15 but
not of HEG19 (Fig. 3). The scale of increase produced
by EGF with HE15 (e.g. 2ERE.TATA, 5.6 6 0.8-fold
increase) is equivalent to that seen with the complete
receptor (2ERE.TATA, 4.8 6 0.7-fold increase). The
product of HE15 does not possess a hormone-binding
site and is therefore unresponsive to both estradiol
and ICI 164384. HEG19 can respond to estradiol, but
there is no further increase seen when EGF is combined with the hormone.
In contrast, CT/IBMX is able to stimulate reporter
gene activity in the presence of HEG19 but not of
HE15 (Fig. 4). The pattern of response to CT/IBMX,
alone or in combination with estradiol or with ICI
164384, in the presence of HEG19 is similar to that
MOL ENDO · 1997
930
Vol 11 No. 7
Fig. 2. ER Dependence of Reporter Gene Response to Estradiol, EGF, and Elevated cAMP in HeLa Cells
HeLa cells were withdrawn from estradiol and transferred to serum-free medium before being transfected with reporter plasmid
DNA with or without ER plasmid (HEG0 or HE0) DNA as indicated. All cells were cotransfected with pSV-b-galactosidase plasmid
DNA. After 24 h, cells were transferred to serum-free medium with additions (E, EGF, CT/IBMX) or without addition (C) as
indicated. Cells were harvested after 24 h of treatment and assayed for CAT and b-galactosidase activity. CAT activity was
normalized relative to thr b-galactosidase activity and is expressed as a percentage of the activity in HEGO-transfected,
estradiol-treated cells. Results are the means 6 SD of three experiments.
seen with HEG0. The response to estradiol in the
presence of HEG19 is reduced (e.g. ERE.VIT, 6.1-fold
increase) compared with that seen with HEG0 (11.1fold increase), but CT/IBMX in combination with estradiol still produces a synergistic response with
ERE.VIT and 2ERE.TATA reporters, although the
ERE.TK response is merely additive (data not presented). Cells transfected with HEG19 were rigorously
depleted of estrogen before they were treated with
CT/IBMX, making it unlikely that the response seen in
the absence of exogenous estradiol involves an interaction between cAMP and ligand-occupied receptor.
The simultaneous presence of HE15 and HEG19
enables the ERE.VIT and ERE.TATA reporter genes to
respond to both estradiol and EGF (Fig. 5a). In addition, the response to EGF is now inhibited by ICI
164384, indicating that the two truncated receptors
interact with each other. Phosphorylation of Ser118 has
been shown to be necessary for full activation of AF-1
(18) and to occur via mitogen-activated protein kinase
in EGF-stimulated cells (19, 20). The HE457 mutant ER
cannot be phosphorylated at this residue and does not
support stimulation of the ERE.VIT or 2ERE.TATA reporter genes by EGF, although their responsiveness to
estradiol and to increased cAMP is retained (Fig. 5b).
The differences in their responses to mutations in
the ER indicate that the two ligand-independent acti-
vators of the ER, EGF and cAMP, achieve their effects
by different mechanisms. We investigated further the
involvement of protein kinases in these two mechanisms. The compound N-[2-((p-bromocinnamyl)amino)ethyl]-5-isoquinolinesulfonamide, 2 HCl (H-89) selectively inhibits cAMP-activated protein kinase
[protein kinase A (PKA)] (21) whereas bisindolylmaleimide I (BIMD) is a highly selective inhibitor of protein
kinase C (22). The response to cAMP of the ERE.VIT
reporter gene, cotransfected into HeLa cells with the
HEG0 expression vector, is abolished by H-89 but is
unaffected by the presence of BIMD (Fig. 6). Interestingly, the response of the reporter gene to estradiol or
to EGF is unaffected by either of these protein kinase
inhibitors. The involvement of PKA in the activation of
the ER by cAMP is supported by the results shown in
Fig. 7. HeLa cells were cotransfected with the ERE.VIT
reporter gene, with the HEG0 ER expression vector
and, as indicated, with an expression vector coding for
either the a- or the b-isoform of the human PKA catalytic subunit (23). Overexpression of PKA alone had
no detectable effect upon reporter gene expression
but, in the presence of estradiol, there was a nearly
4-fold increase in the response above that seen with
estradiol alone. Both PKAa and PKAb produced a
synergistic increase, equivalent to that seen upon elevation of cellular cAMP levels with CT/IBMX. The
Ligand-Independent Activation of ER
931
Fig. 3. The Ability of Carboxyl-Terminal, but not Amino-Terminal, Truncated ER to Support EGF Stimulation of Reporter Gene
Expression in HeLa Cells
HeLa cells were transfected with the indicated reporter plasmid DNA plus expression vector DNA containing either HE15- or
HEG19-truncated ER cDNA. Experiments were conducted as described for Fig. 2 (ICI, antiestrogen ICI 164384). CAT activity is
expressed as a percentage of the activity measured in HEG0-transfected, estradiol-treated cells. Results are the mean 6 SD of
three separate experiments.
effects of PKA upon ERE.VIT expression were abolished by the presence of ICI 164384, indicating their
requirement for ER.
Because responses of transfected estrogen-induced reporter genes have been reported to be cellspecific (6–8), we repeated some of our experiments
using COS-1 cells from monkey kidney. The full-length
ER (HEG0) supported a response to estradiol comparable to that seen in HeLa cells (Fig. 8). Neither reporter
gene showed a significant response to EGF alone,
although the CAT activity with EGF 1 estradiol
(ERE.VIT, 121 6 4%; ERE.TK, 125 6 5%) was marginally higher than that with estradiol alone. Both reporter genes responded to CT/IBMX, but only ERE.VIT
showed a synergistic response when CT/IBMX was
combined with estradiol (CT/IBMX, 2-fold increase;
estradiol, 14-fold increase; estradiol 1 CT/IBMX 36fold increase).
In the presence of HE15, ERE.VIT showed a pattern
of responses similar to that seen in HeLa cells (Fig. 9),
i.e. induction by EGF but not by estradiol or CT/IBMX.
However, ERE.TK responded to EGF and failed to
respond to estradiol, but also showed a clear response to CT/IBMX. ERE.VIT and ERE.TK, in the presence of HEG19, reproduced the pattern of responses
to the three inducers that was seen in HeLa cells, i.e.
induction by estradiol and by CT/IBMX, but only
ERE.TK showed a synergistic response to their
simultaneous presence.
DISCUSSION
Our results demonstrate that the expression of estrogen-responsive reporter genes may also be stimulated
by a polypeptide growth factor, EGF, and by elevated
cAMP, in the absence of exogenous estradiol and in a
manner that is dependent upon the presence of the
ER. The simultaneous presence of an optimally inducing concentration of estradiol (24) and either EGF or
CT/IBMX results in a level of reporter gene expression
that is higher than that seen with estradiol alone. This
suggests that the mechanism of ER activation by EGF
or cAMP differs from that used by estradiol. Furthermore, the fact that EGF 1 estradiol causes an additive
increase whereas CT/IBMX 1 estradiol causes a synergistic increase suggests that the two non-ligand activators differ from each other in their mechanisms of
action.
This latter conclusion is supported by our experiments using truncated versions of the ER, which demonstrate that EGF and cAMP require different regions
of the receptor to achieve their effects upon transcription. The fact that EGF can cause stimulation of transcription in the absence of the carboxyl-terminal AF-2
transactivation domain suggests that its effects are
mediated via AF-1. EGF can stimulate reporter gene
transcription via HE15, a truncated version of the ER
that completely lacks the ligand-binding domain of the
MOL ENDO · 1997
932
Vol 11 No. 7
Fig. 4. The Ability of Amino-Terminal, but not Carboxyl-Terminal, Truncated ER to Support cAMP Stimulation of Reporter Gene
Expression in HeLa Cells
HeLa cells were transfected with the indicated reporter plasmid DNA plus expression vector DNA containing either HE15- or
HEG19-truncated ER cDNA. Experiments were conducted as described for Fig. 2. CAT activity is expressed as a percentage of
the activity measured in HEG0-transfected, estradiol-treated cells. Results are the means 6 SD of three separate experiments.
receptor, clearly establishing that EGF does not simply
enhance the action of estradiol, but is able to activate
the ER de novo. On the other hand, elevated cellular
cAMP levels, produced by treating cells with CT/IBMX,
appear to act by increasing the transactivation activity
of AF-2, located in the carboxyl-terminal half (E domain) of the receptor. In this connection it is interesting
to note that the only difference we observed between
the activity of the wild type receptor (HEG0) and the
receptor with a point mutation in the E domain (HE0) is
that the latter showed a diminished response to CT/
IBMX (Fig. 2). Because the HEG19 construct retains
the ligand-binding domain of the receptor, we cannot
determine whether or not the cAMP effect requires the
occupation of the LBD by an agonistic ligand. The fact
that stimulation of transcription by CT/IBMX can be
seen in cells that have been rigorously depleted of
estrogen makes this unlikely.
The role played by phosphorylation of the ER in the
ability of the receptor to stimulate transcription is unclear. The ER protein is phosphorylated at a basal level
in the absence of estradiol, and numerous reports (e.g.
Refs. 18, 25, and 26) have shown that the level of
phosphorylation is increased upon hormone binding.
The identity of the amino acids phosphorylated, their
relative levels of phosphorylation, and the functional
consequences of their phosphorylation are, however,
the subjects of conflicting reports (18, 25–27). Never-
theless, the fact that ER phosphorylation is also increased, at apparently identical sites, by antiestrogen
binding (18, 25) indicates that ligand-induced transactivation activity of the receptor is not determined solely
by its level of phosphorylation. The evidence we
present here supports the idea that ligand-independent activation of the ER requires phosphorylation of
the receptor protein at sites that differ with the identity
of the activator. Ser118, which is situated within the
AF-1 domain (28), has been identified as a major site of
ligand-stimulated phosphorylation (18, 26) and has
also been shown to be phosphorylated by mitogenactivated protein kinase in response to EGF stimulation of cells (19, 20). Conversion of this residue to a
nonphosphorylatable alanine (HE457) abolishes activation of the ER by EGF (Fig. 5b). However, HE457 is
still activated, albeit to a reduced extent, by estradiol
and by elevated cAMP. Elevation of cellular cAMP has
been reported to stimulate ER phosphorylation at a
site(s) outside the A/B domain and distinct from those
responsive to ligand binding (25). We show that inhibition of cAMP-stimulated protein kinase (PKA), by
treating cells with the specific inhibitor H-89, prevents
stimulation of reporter gene activity by CT/IBMX and,
most strikingly, prevents the synergism between estradiol and CT/IBMX (Fig. 6). The involvement of PKA
in the activation of ER by elevated cAMP is also supported by our observation that increasing kinase ac-
Ligand-Independent Activation of ER
933
Fig. 5. Interaction between Amino-Terminal and Carboxyl-Terminal Truncated ERs in HeLa Cells and the Effect of a Point
Mutation in AF-1 of the ER upon the Stimulation of Reporter Gene Expression
a, HeLa cells were transfected with both HE15- and HEG19-truncated ER plasmid DNA plus the indicated reporter plasmid
DNA. b, HeLa cells were transfected with HE457-mutated ER plasmid DNA plus the indicated reporter plasmid DNA. Experiments
were conducted as described for Fig. 2. CAT activity is expressed as a percentage of the activity measured in estradiol-treated
cells. Results are the mean 6 SD of three separate experiments.
tivity by overexpression of the PKA catalytic subunit
duplicates the synergistic response to estradiol seen
with CT/IBMX treatment (Fig. 7). This stimulation of
reporter gene expression by elevated PKA level was
prevented by the antiestrogen, ICI 164384, indicating
the requirement for ER.
ICI 164384 is termed a “pure” antiestrogen (15) because, unlike non-steroid antiestrogens such as tamoxifen, it does not exhibit any agonist activity but
inhibits both the AF-1 and the AF-2 transactivation
functions of the receptor (29). The mechanism by
which ICI 164384 occupation of the ligand-binding
domain, in the carboxy-terminal half of the receptor,
may interfere with the activity of the amino-terminal
AF-1 function is unclear. Our experiments (Fig. 5a),
in which both truncated versions of the ER
(HE151HEG19) were simultaneously expressed in
HeLa cells, show that the AF-1 domain and the ligandbinding domain do not need to be situated in the same
molecule for inhibition to occur. ICI 164384 was able
to inhibit the action of EGF, previously shown to operate by stimulating AF-1 but not AF-2, indicating that
the two truncated receptors interacted with each other
rather than acting independently. The mechanism of
this intermolecular interaction is unclear. Both HE15
and HEG19 possess DNA-binding domains so that
they may be capable of binding to the ERE as a heterodimer, and the association of HEG19, in an inactive
configuration, with HE15 may inactivate HE15 also.
Both AF-1 and AF-2 contribute to the overall transactivation activity of the ligand-occupied ER, but the
nature of their contributions differs. Truncated receptors have shown that AF-1 can exhibit transactivation
in the absence of estrogen binding to the ER whereas
AF-2 activity is dependent upon hormone binding (6,
7). We have now demonstrated that AF-1 activity, although ligand-independent, is not constitutive but
may be indirectly regulated by signal molecules that
bind to plasma membrane receptors. The relative contributions of AF-1 and AF-2 to gene activation are
markedly influenced both by the structure of the target
gene promoter and by the cell type containing the
target gene (6–8). Differences in their extent of response were observed between the reporter genes,
although the pattern of response to the inducers was
similar for all three reporter gene constructs. However,
it is difficult to detect any consistent pattern in the
relative responsiveness of the reporter genes. In general, the ERE.VIT reporter gene, which has the largest
number of copies of the ERE and the most complex
promoter, gave the largest response, but there were
exceptions to this, e.g. with HEG19 in COS-1 cells
(Fig. 9). We assume that these differences between
promoters arise from differences in requirements for
possible intermediary factors and differences in the
availability of these molecules in different cell types
(30). We compared the responses of the ERE.VIT and
ERE.TK reporter genes in COS-1 cells with those in
MOL ENDO · 1997
934
Vol 11 No. 7
Fig. 6. The Ability of Inhibitors of Protein Kinases to Influence the Induction of Reporter Gene Expression in HeLa Cells
HeLa cells were transfected with HEG0 ER plasmid DNA, ERE.VIT reporter plasmid DNA, and pSV-b-galactosidase plasmid
DNA. Experiments were conducted as described for Fig. 2. Cells were treated for 24 h with the indicated agents, including H-89
(20 mM, PKA inhibitor) or BIMD 1 (10 mM, protein kinase C inhibitor). CAT activity is expressed as a percentage of the activity
measured in estradiol-treated cells. Results are the mean 6 SD of three separate experiments.
Fig. 7. The Influence of Overexpression of PKA Catalytic Subunit upon the Induction of Reporter Gene Expression in HeLa Cells
HeLa cells were transfected with HEG0 ER plasmid DNA, ERE.VIT reporter plasmid DNA, pSVb-galactosidase plasmid DNA
and, where indicated, with either pCaEV (PKAa) or pCbEV (PKAb) expression plasmid DNA. Experiments were conducted as
described for Fig. 2. Cells were treated for 24 h with the indicated agents in medium containing 80 mM ZnSO4. CAT activity is
expressed as a percentage of the activity measured in estradiol-treated cells. Results are the mean 6 SD of three separate
experiments.
Ligand-Independent Activation of ER
935
Fig. 8. The Ability of Estradiol, EGF, and cAMP to Stimulate Reporter Gene Expression in the Presence of ER in COS-1 Cells
COS-1 cells were withdrawn from estradiol and transferred to serum-free medium before being transfected with reporter
plasmid DNA and ER plasmid (HEG0) DNA as indicated. Experiments were conducted as described for Fig. 2. CAT activity is
expressed as a percentage of the activity measured in estradiol-treated cells. Results are the mean 6 SD of three separate
experiments.
HeLa cells. The only significant difference between the
results with the two cell lines was that ERE.TK was
able to respond to CT/IBMX in the presence of HE15,
in COS-1 cells. Apparently, in the presence of certain
promoters and in a cooperative cellular context, cAMP
may also work through AF-1. A potential PKA phosphorylation site does exist within the HE15 sequence
at Ser236 (25).
Our results show that the ER can be stimulated to
activate transcription by EGF and cAMP as well as by
binding estradiol. Transcriptional activation by estradiol is believed to involve both the AF-1 and AF-2
transactivation functions. Our experiments indicate
that, at least in certain promoter/cell type contexts,
EGF and cAMP effects each involve only one transactivation function: AF-1 for EGF and AF-2 for cAMP.
The fact that addition of either EGF or CT/IBMX in the
presence of a saturating concentration of estradiol
results in a further increase in reporter gene expression indicates that there are basic differences between
the mechanisms of ligand-independent and liganddependent activation. We have presented evidence
that specific protein kinases are involved in the mechanisms triggered by the two non-ligand activators of
the ER. It is increasingly recognized that the many
intracellular signal transduction pathways in eukaryotic cells do not operate independently of each other
but that cross-talk between pathways is possible and
potentially important. It appears that the ER provides a
site for the integration of three well recognized signal
transduction pathways, i.e. those of steroid hormones,
cAMP, and tyrosine kinase receptors.
MATERIALS AND METHODS
Chemicals and Materials
Tissue culture medium, newborn calf serum (NBCS), FCS,
antibiotics, and trypsin-EDTA were purchased from GIBCO
BRL (Paisley, Scotland). 17b-Estradiol, insulin, CT, IBMX,
EGF (human, recombinant), proteinase K, and vanadyl complex were obtained from Sigma (Poole, England). ICI 164384
was kindly provided by Dr. A. E. Wakeling, Zeneca Pharmaceuticals (Macclesfield, England). H-89 and BIMD were obtained from Calbiochem-Novabiochem (U.K.) Ltd (Nottingham, England). Other reagents were obtained from
Boehringer Mannheim UK (Lewes, England).
The ER plasmids, reporter constructs, and PKA plasmids
were obtained from the laboratories in which they were constructed. ER derivatives, HE0 (4), HEG0, HEG19 (17), and
HE457 (18) are contained within the eukaryotic expression
vector pSG5, and HE15 is contained in pKCR2 (4). pCaEV
and pCbEV consist of cDNAs coding for the a- and b-isoforms of the PKA catalytic subunit, contained within the Zem3
vector in which transcription is driven by the mouse metallothionein promoter (23). ERE.VIT is derived from the Xenopus laevis vitellogenin B1 gene (31). 2ERE.TATA is an entirely
synthetic enhancer/promoter sequence (31). ERE.TK (pERE
BLCAT) consists of a consensus ERE linked to the promoter
sequence (nucleotides 2105 to 151) of the herpes simplex
virus, thymidine kinase gene (6).
MOL ENDO · 1997
936
Vol 11 No. 7
Fig. 9. The Ability of Truncated ERs to Support the Stimulation of Reporter Gene Expression by Estradiol, EGF, and cAMP in
COS-1 Cells
COS-1 cells were withdrawn from estradiol and transferred to serum-free medium before being transfected with reporter
plasmid DNA and either HE15- or HEG19-truncated ER plasmid DNA, as indicated. Experiments were conducted as described
for Fig. 2. CAT activity is expressed as a percentage of the activity measured in HEG0-transfected, estradiol-treated cells. Results
are the mean 6 SD of three separate experiments.
Cell Culture and Transient Transfections
HeLa cells and COS-1 cells were obtained from the European
Collection of Animal Cell Culture (Porton Down, U.K). Both
cell lines were cultured in MEM medium (GIBCO BRL) containing 100 mg/ml penicillin (GIBCO), 100 mg/ml streptomycin
(GIBCO), and 10% (vol/vol) FCS (GIBCO). Cells were depleted of estrogen by culture for 6 days in phenol red-free
RPMI 1640 medium (GIBCO), supplemented with 5% (vol/vol)
dextran/charcoal-treated NBCS (32). Cells were further cultured for 1 day in medium supplemented with 1% (vol/vol)
dextran/charcoal-treated NBCS followed by 1 day in serumfree medium [phenol red-free RPMI 1640 supplemented with:
10 mg/ml transferin; 10 ng/ml sodium selenite; 1% (wt/vol)
glutamine; 0.2% (wt/vol) BSA]. Cells were then harvested and
seeded, in multiwell plates, at a density of 2 3 105 cells per
35-mm well, in serum-free medium. After 24 h, cells were
transfected with DNA [300 ng ER plasmid (HEG0, HE0, HE15,
HEG19 or HE457), 1000 ng each of 2ERE.TATA, ERE.VIT, or
ERE.TK plasmid, 20 ng pCaEV or pCbEV plasmid, 700 ng
pSV-b-galactosidase plasmid (Promega, Madison, WI)] using
the LipofectAMINE reagent (Life Technologies, Gaithersburg,
MD) according to the manufacturer’s protocol. After 5 h an
equal volume of serum-free medium was added to each well.
After a further 19 h the medium was changed for fresh serumfree medium containing the inducing agents: 1028 M estradiol
(E); 100 ng/ml EGF; 1 mg/ml CT, 1024 M IBMX; 1026 M ICI
164384 (ICI); or without addition (C). For transfections involving the pCaEV and pCbEV vectors, the culture medium was
supplemented with 80 mM ZnSO4. Cells were incubated for
24 h and then were harvested and CAT (liquid scintillation
method) and b-galactosidase activities were assayed in
whole cell extracts using assay kits (Promega) according to
the manufacturer’s protocols. CAT activity results were normalized relative to the b-galactosidase activity, to correct for
differences in efficiency of transfection, and are expressed as
a percentage of the estradiol-induced level, as indicated.
Results are the mean 6 SD of at least three separate
experiments.
Acknowledgments
We wish to thank P. Chambon for the gift of HEG0, HE0,
HE15, HEG19, and HE457; G. S. McKnight for the gift of
pCaEV and pCbEV; D. J. Shapiro for the gift of ERE.VIT and
2ERE.TATA; M. G. Parker for the gift of ERE.TK; and A. E.
Wakeling for the gift of ICI 164384.
Received July 19, 1996. Re-revision received and accepted March 11, 1997.
Address requests for reprints to: Dr. C. D. Green, School of
Biological Sciences, Life Sciences Building, University of
Liverpool, P.O. Box 147, Liverpool, L69 3BX, U.K.
This work was supported by the Association for International Cancer Research.
REFERENCES
1. Evans RM 1988 The steroid and thyroid hormone receptor superfamily. Science 240:889–895
2. Tsai M-J, O’Malley BW 1994 Molecular mechanisms of
action of steroid/thyroid receptor superfamily members.
Annu Rev Biochem 63:451–486
3. Beato M, Herrlich P, Schutz G 1995 Steroid hormone
Ligand-Independent Activation of ER
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
receptors: many actors in search of a plot. Cell
83:851–857
Kumar V, Green S, Stack G, Berry M, Jin J-R, Chambon
P 1987 Functional domains of the human estrogen receptor. Cell 51:941–951
Parker MG, Arbuckle N, Dauvois S, Danielian P, White R
1993 Structure and function of the estrogen receptor.
Ann NY Acad Sci 684:119–126
Lees JA, Fawell SE, Parker MG 1989 Identification of two
transactivation domains in the mouse oestrogen receptor. Nucleic Acids Res 17:5477–5488
Tora L, White J, Brou C, Tasset D, Webster N, Scheer E,
Chambon P 1989 The human estrogen receptor has two
independent nonacidic transcriptional activation functions. Cell 59:477–487
Tzukerman MT, Esty A, Santiso-Mere D, Danielian P,
Parker MG, Stein RB, Pike WJ, McDonnell DP 1994 Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions.
Mol Endocrinol 8:21–30
Ignar-Trowbridge DM, Teng CT, Ross KA, Parker MG,
Korach KS, McLachlan JA 1993 Peptide growth factors
elicit estrogen receptor-dependent transcriptional activation of an estrogen-responsive element. Mol Endocrinol
7:992–998
Couse JF, Curtis SW, Washburn TF, Eddy EM, Schomberg
DW, Korach KS 1995 Disruption of the mouse oestrogen
receptor gene: resulting phenotypes and experimental
findings. Biochem Soc Trans 23:929–935
Chalbos D, Philips A, Galtier F, Rochefort H 1993 Synthetic antiestrogens modulate induction of pS2 and cathepsin-D messenger ribonucleic acid by growth factors
and adenosine 39,59-monophosphate in MCF-7 cells. Endocrinology 133:571–576
Smith CL, Conneely OM, O’Malley BW 1993 Modulation
of the ligand- independent activation of the human estrogen receptor by hormone and antihormone. Proc Natl
Acad Sci USA 90:6120–6124
Cho H, Aronica SM, Katzenellenbogen BS 1994 Regulation of progesterone receptor gene expression in MCF-7
breast cancer cells: a comparison of the effects of cyclic
adenosine 39,59-monophosphate, estradiol, insulin-like
growth factor-1 and serum factors. Endocrinology
134:658–664
Ince BA, Montano MM, Katzenellenbogen BS 1994 Activation of transcriptionally inactive human estrogen receptors by cyclic adenosine 39,59- monophosphate and
ligands including antiestrogens. Mol Endocrinol
8:1397–1406
Wakeling AE, Bowler J 1988 Novel antioestrogens without
partial agonist activity. J Steroid Biochem 31:645–653
Telfer JF, Green CD 1993 Placental alkaline phosphatase
activity is inversely related to cell growth rate in HeLaS3
cervical cancer cells. FEBS Lett 329:238–244
Tora L, Mullick A, Metzger D, Ponglikitmongkol M, Park
I, Chambon P 1989 The cloned human oestrogen receptor contains a mutation which alters its hormone binding
properties. EMBO J 8:1981–1986
Ali S, Metzger D, Bornert J-M, Chambon P 1993 Modulation of transcriptional activation by ligand-dependent
phosphorylation of the human oestrogen receptor A/B
region. EMBO J 12:1153–1160
Kato S, Endoh H, Masuhiro Y, Kitamoto T, Uchiyama S,
Sasaki H, Masushige S, Gotoh Y, Nishida E, Kawashima
937
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
H, Metzger D, Chambon P 1995 Activation of the estrogen receptor through phosphorylation by mitogen-activated protein kinase. Science 270:1491–1494
Bunone G, Briand P-A, Miksicek RJ, Picard D 1996 Activation of the unliganded estrogen receptor by EGF involves the MAP kinase pathway and direct phosphorylation. EMBO J 15:2174–2183
Chijiura T, Mishima A, Hagiwara M, Sano M, Hayashi K,
Inoue T, Naito K, Tashioka T, Hidaka H 1990 Inhibition of
forskolin-induced neurite outgrowth and protein phosphorylation by a newly synthesised selective inhibitor of
cyclic AMP-dependent protein kinase, N-[2-(p-bromocinnamylamino) ethyl]-5-isoquinolinesulfonamide (H-98),
of PC12D pheochromocytoma cells. J Biol Chem 265:
5267–5272
Toullec D, Pianetti P, Coste H, Bellevergue P, GrandPerret T, Ajakane M, Baudet V, Boissin P, Boursier E,
Loriolle F, Duhamel L, Charon D, Kirilovsky J 1991 The
bisindolylmaleimide GF109203X is a potent and selective
inhibitor of protein kinase C. J Biol Chem 266:15771–
15781
Uhler MD, McKnight GS 1987 Expression of cDNAs for
two isoforms of the catalytic subunit of the cAMP-dependent protein kinase. J Biol Chem 262:15202–15207
El-Tanani MKK, Green CD 1995 Oestrogen-induced
genes, pLIV-1 and pS2, respond divergently to other
steroid hormones in MCF-7 cells. Mol Cell Endocrinol
111:75–81
Le Goff P, Montano MM, Schodin DJ, Katzenellenbogen
BS 1994 Phosphorylation of the human estrogen receptor: identification of hormone-regulated sites and examination of their influence on transcriptional activity. J Biol
Chem 269:4458–4466
Joel PB, Traish AM, Lannigan DA 1995 Estradiol and
phorbol ester cause phosphorylation of serine 118 in the
human estrogen receptor. Mol Endocrinol 9:1041-1052
Arnold SF, Varojeikina DP, Notides AC 1995 Phosphorylation of tyrosine 537 on the human estrogen receptor is
required for binding to an estrogen response element.
J Biol Chem 270:30205–30212
Metzger D, Ali S, Bornert J-M, Chambon P 1995 Characterisation of the amino-terminal transcriptional activation function of the human estrogen receptor in animal
and yeast cells. J Biol Chem 270:9535–9542
Metzger D, Berry M, Ali S, Chambon P 1995 Effect of
antagonists on DNA-binding properties of the human
estrogen receptor in vitro and in vivo. Mol Endocrinol
9:579–591
Le Douarin B, vom Baur E, Zechel C, Heery D, Heine M,
Vivat V, Gronemeyer H, Losson R, Chambon P 1996
Ligand-dependent interaction of nuclear receptors with
potential transcriptional intermediary factors (mediators).
Philos Trans R Soc Lond [Biol] 351:569–578
Chang T-C, Nardulli AM, Lew D, Shapiro DJ 1992 The
role of estrogen response elements in expression of the
Xenopus laevis vitellogenin B1 gene. Mol Endocrinol
6:346–354
Darbre PD, Yates J, Curtis S, King RJB 1983 Effect of
estradiol on human breast cancer cells in culture. Cancer
Res 43:349–354
Chang T-C, Shapiro DJ 1990 An NF1-related vitellogenin
activator element mediates transcription from the estrogen-regulated Xenopus laevis vitellogenin promoter.
J Biol Chem 265:8176–8182