| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Neuroendocrine Unit Massachusetts General Hospital and Department of Medicine Harvard Medical School Boston, Massachusetts 02114
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(ER) isoform on estrogen
(E2)-mediated gene activation in U2-OS
osteosarcoma cells. 
5ER, an alternatively spliced ER variant
lacking exon 5, is coexpressed with normal ER in several
E2-responsive neoplastic tissues. However, the
potential interactions of 5ER with normal ER have not been
functionally characterized. 5ER encodes the hormone-independent
trans-activating function (AF-1), as well as the
constitutive receptor dimerization and DNA-binding domains. It is
generated by an alternate splice event that omits exon 5 and alters the
reading frame of the resulting mRNA. The 5ER protein is
prematurely truncated and lacks the majority of the hormone-binding
and activating function-2 (AF-2) domains. When 5ER mammalian
expression vector was transfected alone in human ER/ERß-negative
osteosarcoma U2-OS cells, it had no effect on either basal or
E2-mediated EREtk81Luc reporter transcriptional
activity, while transfected cells expressing control normal ER
increased EREtk81Luc activity up to 20-fold in response to 10
nM E2. However, when
5ER was cotransfected with normal ER, both basal and
E2-stimulated EREtk81Luc reporter activation
were increased approximately 500% over levels observed when cells were
transfected with ER alone. Similar effects of 5ER and normal
ER coexpression were observed using an
E2-responsive human C3 promoter/luciferase
reporter construct. The effects of 5ER on normal ER were
further assessed in U2-OS cells stably transfected with normal ER.
Transfection of increasing amounts of 5ER expression vector into
[ER+]OS cells resulted in potentiation of
E2-stimulated ERELuc activity in a synergistic,
dose-dependent manner. Moreover, coexpression of 5ER in
[ER+]OS cells improved E2 sensitivity
100-fold over cells expressing ER alone. Proliferation rates of
stable U2-OS cell lines expressing 5ER were significantly
increased (P < 0.05), with cell doubling times
reduced from 35 h in control parental U2-OS cells to 28 h in
[5ER]OS cells. However, growth rates were not affected by
either E2 or tamoxifen treatment.
Electromobility shift/supershift assays using nuclear extracts of U2-OS
cells stably transfected with ER and 5ER confirmed the
constitutive binding of 5ER and ER protein to
estrogen-response element (ERE) sequence independent of
E2 and also showed an increase in
5ER/ER-ERE complexes with E2
treatment. These data are consistent with interactive effects of normal
ER and 5ER on transcription from classic ERE gene promoters.
5ER appears to therefore act as a dominant positive receptor that
increases both basal and E2-stimulated gene
transactivation of normal ER. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and -ß (ER and ERß), ligand-activated
transcription factors that are members of the steroid receptor
superfamily (3, 4). The genomic structure of the ER gene reveals
strong homology to viral v-erbA, suggesting that ER is a
cellular homolog of this oncogene (3). ER is encoded by eight exons
within a genomic locus of greater than 140 kb (5). It has been well
documented that ER isoform variants generated by alternative
splicing of ER heteronuclear RNA are coexpressed in a number of
human normal and neoplastic tissues (6, 7, 8, 9, 10, 11, 12, 13). Thus, ER gene
expression and alternative splicing have been postulated to create a
heterogeneous population of ER isoforms with differential
transcriptional activity, which may help to potentiate the diverse
action of E2 through a single gene.
The ER protein is composed of several structural domains, each of
which has a unique function in ligand binding, gene promoter
activation, and association with other members of the general
transcriptional apparatus (14). The A/B domain has a ligand-independent
gene activation function (AF-1) and is encoded by exon 1. It has been
shown to be important for stimulating transcription from certain
E2-responsive genes such as pS2 (14), c-fos
(15), and C3 (16). In addition, it is also critical for growth factor
interactions with ER-signaling pathways both in yeast and mammalian
cells (17). However, because the A/B domain cannot bind DNA directly,
it has been hypothesized to activate target genes by associating with
components of the core transcriptional machinery, such as TFIID and
other coactivators/repressors (18). The ER DNA-binding domain (DBD)
lies within the C region and is encoded by exons 2 and 3. The DBD is
composed of two type II zinc (Zn) finger motifs that have been shown by
structure/function studies to be directly responsible for DNA promoter
sequence recognition (19). The D region (or variable hinge region) is
thought to allow the ER to alter conformation and is encoded by exon
4. This ER domain may potentiate much of the allosteric regulation
of the receptor after ligand binding (14). It also contains a
constitutive nuclear localization signal as well as sequences required
for dimerization of the ER. Finally, the COOH-terminal E region is
encoded by exons 5–8. It is functionally complex and is the most
characterized domain in terms of structure and function (20, 21, 22). The E
region contains protein sequences important for 1) heat-shock protein
association in the cytoplasm, 2) nuclear localization, 3)
ligand-dependent receptor dimerization and the AF-2 gene activation
function, and 4) E2 and antiestrogen ligand binding.
This study focuses on the cooperative effects of human 5ER. This
variant encodes the A/B domain as well as the C and D domains critical
for binding estrogen-response elements (EREs), receptor dimerization,
and nuclear localization, but lacks the hormone-binding domain. It is
generated by an exon 5 splice deletion, and the subsequent fusion of
exons 4 and 6 results in an immediate frame shift, a short novel
carboxyl terminus (GTRQNV), and termination codon. The exon-5 ER
spliced variant 5ER has been shown by others to have
approximately 10–15% constitutive transcriptional activity of normal
ER when expressed alone in yeast (23). However, little is known as
to their potential cooperative effects when coexpressed with normal
ER. We and others have shown that in E2-sensitive human
tissues, 5ER variant isoform was typically found to be
coexpressed with normal receptor. Both tamoxifen-resistant and primary
breast tumors express 5ER along with normal ER. 5ER
expression was elevated in ER+ tumors that were tamoxifen-resistant
and in ER- tumors that expressed the E2-responsive
markers, PgR (progesterone receptor) and pS2 (24). In pituitary tumors,
5ER was found to be tumor specific and coexpressed with normal
ER only in prolactinomas and gonadotroph tumors, but not in normal
pituitary or other pituitary tumor phenotypes (13). Therefore, we
hypothesized that 5ER may interact with normal ER and may play
a role in tumor pathogenesis
The present study examined the hypothesis that the 5ER
tumor-specific splice variant is capable of regulating
E2-responsive genes, which, in turn, control cellular
phenotype and growth. These experiments investigated 1) the
transcriptional effects of coexpression of 5ER and normal ER
on E2-responsive gene promoter/reporter constructs, 2) the
ability of 5ER and normal ER to stably bind ERE complexes in
electrophoretic mobility shift/supershift assays (EMSAs) using nuclear
extracts from stable cell lines harboring either normal ER or
5ER, and 3) the effects of 5ER on cellular proliferation
rates in stably transfected human U2-OS osteosarcoma cell lines.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and 5ER and 5ER full-length cDNA were cloned by RT-PCR
from pooled human gonadotroph and lactotroph tumor mRNA. These
pituitary tumor subtypes have been previously shown to express the
5ER variant in a tumor-specific manner along with full-length
ER (13). Figure 1A
shows the
exon/intron structure and functional domains of normal ER as well as
5ER. 5ER is created by an RNA splice omission of exon 5,
causing a fusion of exons 4 and 6, which acts to shift the open-reading
frame and creates a truncated carboxy terminus. As a result of this
altered splice event, 5ER lacks almost the entire ligand-binding
and AF-2 region and cannot bind E2. Figure 1B shows the
expected in vitro translation protein products from normal
human ER- and 5ER-cloned cDNAs.
[35S]methionine-labeled protein products from ER and
5ER cDNAs migrated at the expected sizes of 67 and 41 kDa,
respectively.
|
and 5ER cDNAs were ligated into the
pBKCMV expression plasmids to study their action in mammalian cells.
U2-OS cells were stably transfected with either normal ER or
5ER to generate cell lines which harbor and constitutively
express each ER isoform. Geneticin (G418)-resistant clonal U2-OS
cell lines were isolated and human ER isoform expression was
examined by immunocytochemistry using a human ER-specific antibody
which recognized an epitope within the A/B domain (residues 22–43) of
both 5ER and normal ER. Immunocytochemical data for these
stable cell lines are shown in Fig. 2.
Figure 2A shows negative control parental U2-OS cells, which, based on
RT-PCR analyses, do not express any form of ER or ERß. No staining
could be seen in these cells with a specific human ER-1 antibody
directed against the amino terminus of ER, while control in
ER-positive MCF-7 breast cancer cells exhibited demarcated nuclear
staining (data not shown). Figure 2, B and C, shows immunocytochemical
localization of 5ER and ER, respectively, within the
nuclear/perinuclear compartment. The pBKCMV-driven expression of each
ER isoform appeared to be at equivalent levels in the [ER]OS
and [5ER]OS stable cell lines used in this analysis. No
staining was seen in any stable cell line when primary antibody was
omitted or preabsorbed to in vitro translated ER protein
(data not shown). Together, the in vitro
transcription/translation and immunocytochemical data demonstrate the
predicted protein size and cellular protein expression of both ER
and its 5ER splice variant in stably transfected U2-OS cell
lines.
|
5ER Enhances Both Basal and
E2-Stimulated Gene Expression by Normal
ER or ERß as measured by
RT-PCR. We confirmed the reported U2-OS cell estrogen-resistant
phenotype by transiently transfecting the E2-responsive
human EREtk81Luc reporter construct with or without pBKCMV/ER
mammalian cell expression vector. Figure 3 demonstrates that U2-OS cells are
unresponsive to E2 levels as high as 0.1 µM,
with no significant increases in ERE-driven luciferase activity above
control wells. However, when transiently transfected with pBKCMV/ER,
U2-OS cells exhibited significant (P < 0.05)
up-regulation of EREtk81Luc activity at E2 doses as low as
1 nM. Normal ER activation of the reporter construct was
found to be E2 dose-dependent, ranging from 9- to 15-fold
activation with administration of 10-9 to
10-7 M E2, respectively. No
activation was seen at any E2 dose in cells that were
transiently transfected with empty pBKCMV vector. These data confirm
that U2-OS cells lack endogenous ER or ERß with a concomitant
inability to activate ERE transcription. Moreover, the nonresponsive
phenotype can be rescued by transfection of pBKCMV/ER, which enables
U2-OS cells to up-regulate E2-dependent gene transcription
in response to physiological levels of exogenous E2.
|
and 5ER on
E2-responsive gene transcription, U2-OS cells were
transiently transfected with ER and/or 5ER along with
EREtk81Luc, followed by treatment with E2. The results of
these cotransfection studies are shown in Fig. 4A. 5ER had no constitutive or
E2-stimulated effects on EREtk81Luc transcription when
expressed alone in U2-OS cells. However, coexpression of 5ER with
normal ER resulted in significant up-regulation of both basal
(P < 0.01) and E2-stimulated
(P < 0.001) EREtk81Luc activity 4- and 33-fold,
respectively, over basal levels in the control osteosarcoma (OS)
cells, or approximately 5-fold compared with cells transfected with
ER alone. Transfection of 2 µg of ER increased basal and
E2-stimulated luciferase reporter activity approximately
2-fold over levels seen with transfections using 1 µg of ER. The
dose-dependent effect of 5ER on the regulation of EREtk81luc gene
transcription was also examined in stably transfected [ER+]OS
cells (Fig. 4B). Coexpression of 5ER with normal ER
significantly (P < 0.001) increased
E2-mediated trans-activation of the classic ERE
promoter in a dose-dependent manner, up to 90-fold with 3 µg
5ER compared with [ER+]OS cells transfected with empty
pBKCMV vector. Basal levels of EREtk81luc activity also significantly
increased (P < 0.001) in a dose-dependent manner up to
314% of baseline.
|
5ER with normal ER
enhances OS cells sensitivity to lower E2 dose
administration, E2 dose-response curves were conducted in
[ER+]OS cells that were cotransfected with either empty pBKCMV or
cytomegalovirus (CMV)/5ER along with the EREtk81Luc
reporter. Cells were then treated with increasing [10-13
M to 10-7 M] E2 doses
for 24 h, and EREtk81Luc activity was subsequently assayed. Figure 5 shows an E2
dose-response curve in [ER+]OS cells. Cells that coexpressed
5ER were 100-fold more sensitive to E2 administration
than OS cells expressing normal ER alone. Moreover, the magnitude of
the transcriptional response was on average 550% at higher
E2 doses [10-10 M to
10-7 M].
|
. The C3T1luc reporter
exhibited an identical pattern of activation as the heterologous
EREtk81luc reporter in U2-OS cells. Basal levels of C3T1luc activity in
wells cotransfected with both ER and 5ER were enhanced by
5.5-fold over those seen in wells transfected with either ER or
5ER alone. E2-stimulated C3T1luc activity was increased 33-fold
over the basal levels and 8.7-fold over E2-treated cells transfected
with ER alone. In contrast, when a c-fosLuc reporter containing
imperfect ERE sequences was used, both ER and 5ER showed very
little effect on c-fos transcription activity (Fig. 6B),
suggesting that the classic palindromic ERE sequences are necessary for
the cooperative transcriptional effect of ER and 5ER.
|
and 5ER U2-OS
Stable Cell Lines and/or its 5ER protein
variant, gel-shift assays were performed on nuclear extracts from
normal ER and 5ER stable cell lines, and results from these
EMSAs are shown in Fig. 7. There was a
faint, but detectable, band shift of labeled ERE oligonucleotide in
nuclear extracts prepared from parental ER-negative U2-OS cells
(lanes 1 and 2). As expected, nuclear extracts derived from
ER-stably transfected cells showed a marked increase in protein-ERE
complexes with E2 treatment (lanes 3 and 4). Conversely,
5ER nuclear extracts were capable of binding ERE oligonucleotide,
independent of E2 (lanes 5 and 6). In addition, in
ER/5ER mixed extracts, this shifting pattern was increased in
intensity in the presence of E2 (lanes 7 and 8). The specificity of the
gel-shift was demonstrated by competition assays with excess, unlabeled
ERE DNA (Fig. 7, lanes 9–14).
|
and 5ER were carried out to
determine whether the observed EMSA patterns were due to direct
interactions of normal and 5ER variant with ERE complexes. Figure 8 shows the results of these experiments
with nuclear extract from each stable U2-OS cell line. There was no
detectable supershift with either antibody in parental OS extracts,
while both antibodies supershifted ERE DNA in the normal ER
extracts. As expected, the amino terminus antibody supershifted ERE in
the 5ER extracts. However, the carboxy-terminus antibody, which
recognizes an epitope in the hormone-binding domain distal to the
5ER premature termination codon, failed to supershift
5ER/ERE complexes. Finally, normal ER/5ER mixed nuclear
extracts, as expected, displayed supershifted patterns of ER/ERE
complexes with both antibodies.
|
and 5ER Variant5ER variant,
we conducted growth studies of each stable cell line. Stable
transfectant U2-OS cell clones were screened for the expression of
ER and 5ER mRNA by Northern blot analysis (data not shown) and
immunocytochemistry (ICC) using a monoclonal antibody against the A/B
domain common to each ER receptor studied (as shown in Fig. 2
).
Figure 9 shows representative growth
curves of each OS cell line. Cell-doubling times derived from the log
phase of the growth curve are the average of three proliferation
studies. Cell doubling time for parental OS cells was 35 ±
3.5 h (mean ± SD). OS cells stably transfected
with normal ER exhibited slightly slower cell-doubling rates of
39 ± 4.0 h, but the difference was not statistically
significant by two-way ANOVA analysis. However, stable [5ER]OS
cells exhibited a decrease in cell doubling time of 28 ± 1.2
h, a change that was also statistically significant (P
< 0.05) over both parental and [ER+]OS cell lines. Addition of
either 10 nM E2 or 1 µM tamoxifen
to culture medium had no significant effect on the growth rate of OS,
[ER+]OS, and [5ER]OS cell lines (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
5ER
isoform with normal ER on estrogen (E2)-mediated gene
activation in U2-OS osteosarcoma cells. It demonstrates that the human
5ER isoform can markedly enhance the transcriptional effects of
normal ER on E2-responsive genes in this human cell line. 5ER,
an alternatively spliced ER variant lacking exon 5, is coexpressed
with normal ER in several E2-responsive neoplastic
tissues, including primary pituitary (13), endometrial (25), and breast
tumors (26, 27). These studies provide direct, functional evidence
demonstrating that 5ER can markedly up-regulate both basal and
E2-stimulated classical ERE promoter gene activation when
coexpressed with normal ER in vitro. It also details
experiments describing the subcellular distribution, DNA-binding
ability, and proliferative potential of 5ER variant in the human
U2-OS cell line.
As predicted by its structure, the 5ER variant has similarities
to normal ER as well as important differences. It encodes the A/B
domains as well as the C and D domains critical for DNA binding and
constitutive nuclear localization, respectively. EMSA experiments
demonstrate that nuclear extracts from cells stably expressing
5ER can bind ERE sequences in the presence and absence of
E2. Immunocytochemistry utilizing stable cell lines
harboring the 5ER expression vector confirm that the truncated
protein is expressed and at least partially compartmentalized in the
cell nucleus. However, due to an alternate splice event that omits exon
5 and alters the reading frame of the resulting mRNA, the 5ER
protein is prematurely truncated and lacks the majority of the
hormone-binding and activating function-2 (AF-2) domains. Coupled
in vitro transcription/translation of 5ER cDNA
demonstrates that the cloned isoform encodes the expected 41-kDa
protein product lacking the hormone binding and AF-2 domains. Transient
transfection of 5ER demonstrates that it is unable to
trans-activate E2-responsive reporter vectors in
the presence of E2.
Previous studies have conclusively demonstrated that 5ER has
minimal constitutive activity when expressed either in yeast or
mammalian cells (6, 7, 23, 28). Experiments in which 5ER was
transfected with either the EREtk81Luc or C3Luc reporter constructs
confirmed these results. When 5ER cDNA was transfected alone in
human ER/ERß-negative osteosarcoma U2-OS cells, it had no effect
on either basal or E2-mediated EREtk81Luc or C3T1luc
reporter transcriptional activity. However, this cell line clearly
contains the basic transcriptional apparatus to up-regulate
E2-stimulated gene activity, because transfected cells
expressing control normal ER increased EREtk81Luc activity up to
20-fold in response to 10 nM E2. These findings
with normal ER are in agreement with other studies that examine
normal ER activation of E2-responsive gene transcription
in various OS cell lines (29, 30, 31, 32, 33). Our data also confirm previous
studies that show that the 5ER isoform has minimal activity when
expressed alone. However, when 5ER was cotransfected with normal
ER, both basal and E2-stimulated EREtk81Luc reporter
activation were increased approximately 500% over levels observed when
cells were transfected with ER alone. Similar effects of 5ER
and normal ER coexpression were observed using an
E2-responsive human C3 promoter/luciferase reporter
construct.
Because these data were obtained using transient transfection of U2-OS
cells, we chose to extend and confirm these observations in stable cell
lines expressing normal ER. This approach allowed us to
quantitatively assess the impact of 5ER coexpression on normal
ER transactivation. Transfection of increasing amounts of 5ER
expression vector into [ER+]OS cells resulted in potentiation of
both basal and E2-stimulated ERELuc activity in a
synergistic, dose-dependent manner. Collectively, these transient
and stable transfection studies confirm what the structure of 5ER
predicts: When expressed alone in an ER-negative cell, 5ER is
unable to bind E2 and fails to up-regulate either
EREtk81luc or C3T1luc gene activity either in the presence or absence
of E2. However, the unexpected result in this study is that
5ER can act as a dominant positive receptor isoform and
facilitate both basal and E2-stimulated ERE-mediated
transcription of normal ER when coexpressed in U2-OS cells.
The mechanism underlying the observed constitutive and
E2-stimulated transcriptional activation of the classic ERE
promoter by 5ER and normal ER may be due to formation a
5ER/ER heterodimer capable of binding ERE promoter sequences
and activating transcription independent of E2. Several
studies utilizing site-specific mutagenesis of ER in both yeast
expression systems and MCF-7 cells have led to the general hypothesis
that ER activation of gene transcription is facilitated by an
interaction between AF-1 in the A/B domain and AF-2 in E domain
(34, 35, 36). However, both AF-1 and AF-2 can also activate transcription
independently, and each can bind the basal transcription components of
the preinitiation complex directly, notably TFIIB and TFIID, in a
ligand-independent manner in vitro (37, 38). In addition,
5ER encodes a recently described third activation domain, AF2a,
located between residues 282 and 351 of the D region (39). Data from
electromobility shift/supershift assays using nuclear extracts of U2-OS
cells stably transfected with ER and 5ER confirmed the
constitutive binding of 5ER protein to ERE-containing complexes
in the presence and absence of E2. Parallel with
transcriptional studies, EMSA also showed the expected increase in
ER-or ER/5ER-ERE complexes with E2 treatment.
The observed increase in the receptor-ERE bound complex with
E2 treatment has been described as a result of the
formation of ER DBD-ERE complexes with greater stability (19, 40).
These findings suggest the possibility that altered structure of
5ER truncated receptor may promote enhanced binding to ERE
promoters even in the absence of E2. Taken together, the
results from functional transfection studies on the classic ERE
transcription and gel shift assays of direct DNA-protein interactions
in vitro are consistent with both constitutive and
cooperative transcriptional activation of the classic ERE promoter by
the coexpressed 5ER and normal ER.
In this study we found that human osteosarcoma cell lines stably
transfected with a 5ER expression vector exhibit
E2-independent increases in cellular growth rates.
Proliferation rates of stable U2-OS cell lines expressing 5ER
were significantly increased (P < 0.05), with
cell-doubling times reduced from 35 h in control parental U2-OS
cells to 28 h in [5ER]OS cells. Tamoxifen or
E2 treatment had no significant effect on cellular
proliferation rates in any of the U2-OS stable cell lines studied. This
is not unexpected in 5ER stable lines because of a lack of
ligand-binding domain in the truncated receptor. In contrast to our
proliferation data in U2-OS cells, 5ER has been shown to have
little effect on MCF-7 breast cancer cell growth in response to
E2 or tamoxifen (41). These differences in 5ER growth
effects in MCF-7 cells may be confounded by endogenous production of
5ER as well as a number of other ER variants in that cell line
(8). Alternatively, these data may represent cell-specific mechanisms
of 5ER action on cellular proliferation in human osteosarcoma
cells.
In contrast to growth effects seen with 5ER, U2-OS cells stably
transfected with normal ER showed insignificant alterations in
proliferative potential when compared with parental U2-OS cells and
were not affected by treatment with either E2 or tamoxifen.
These findings are consistent with several characterized human
osteosarcoma cell lines which either fail to respond or inhibit growth
in the presence of E2 in vitro. These
proliferative responses may depend, in part, on exogenous ER
expression levels. For example, in the HTB 96 human OS cell line,
E2 caused a growth-inhibitory response in lines that were
stably transfected with ER, when compared with a parental HTB 96 OS
cell line (32). Human SaOS-2 cell lines stably transfected with ER
also exhibit growth inhibition when treated with E2 (29).
However, similar growth studies with HOS TE85 human osteoblastic cells
exhibited no proliferative response to exogenous E2 (30).
Thus, although the proliferative effects of E2 in other
human cell lines derived from E2-sensitive tissue
(i.e. uterus and breast) are well-characterized, the data in
osteoblast-like osteosarcoma cell lines suggest E2 effects
on proliferative potential are limited in this in vitro
cellular system.
It is unclear how the truncated 5ER lacking the ligand-binding
domain can enhance cellular proliferation in stable U2-OS cells.
Several studies have demonstrated that the A/B domain containing AF-1
of the ER is important for peptide growth factor interaction,
independent of E2 (18, 37). Moreover, a
mutagenesis-generated ER isoform lacking much of the hormone-
binding domain E has also been shown to activate c-fos
promoter independent of E2 administration in HeLa cells
(42). Therefore, the naturally occurring 5ER, which encodes only
the A/B (AF-1) and DBD domains, might function to up-regulate growth
factor-induced proliferative responses. One potential hypothesis is
that this isoform is a potent target for peptide growth factor
kinase-signaling pathways that influence its interactions with other
coregulatory proteins. This may result in enhanced cellular growth in
the absence of stimulation of ERE-dependent gene activation, and an
E2-independent growth phenotype.
Studies examining potential interactions of structurally altered ER
isoforms may have important implications for the well described ER
isoform coexpression observed in human E2-sensitive
neoplastic tissues. If ER isoform expression plays a role in growth
regulation of E2-sensitive tumor cells, dysregulation of
mRNA-splicing mechanisms that give rise to these variants in neoplastic
cells may be selective during tumor progression to a more aggressive
phenotype. These data are consistent with interactive effects of normal
ER and 5ER on transcription from classic ERE gene promoters
and suggest that coexpression of these ER isoforms may alter
cellular phenotype, growth, and E2-mediated gene
activation. Given the observed effects of 5ER on normal ER
function, we hypothesize that this 5ER variant may have
pathophysiological consequences in these tissues. Based on these data,
we hypothesize that 5ER potentiates E2 and normal
ER actions on cellular growth, differentiation, and/or neoplastic
progression in human E2-responsive tissues expressing both
5ER and normal ER.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and its variant 5ER were
cloned from human pituitary tumor cDNA into the eukaryotic pBK-CMV
expression vector (Stratagene, La Jolla, CA). Briefly, pooled pituitary
first-strand cDNA obtained from human lactotroph and gonadotroph tumors
by reverse transcription served as template for PCR amplification using
a high-fidelity Pfu DNA polymerase (Stratagene) as
previously described (13). Primer pairs SpeI-ER (5'-gga
cta gtc cat gac cat gac cct CCA-3') and ER4L (5'-ttc gcc cag ttg atc
atg tg-3') were used to amplify the N-terminal portion of ER
(nucleotides 231-1298, GenBank Accession no. X03635) while ER4U
(5'-gcc ccc cat act cta ttc-3') and ER-ClaI (5'-gga tcg
atg cag cag gga tta tct ga-3') were used to amplify the C-terminal part
of ER (nucleotides 1201–2063). PCR products of the appropriate
sizes for normal and variant ER fragments were purified from 1%
agarose gel using GlasPac/GS purification kit (National Scientific, San
Rafael, CA). The N-terminal portion of ER fragment was digested with
SpeI and HindIII while the C-terminal portion of
ER and 5ER fragments were digested with HindIII and
ClaI before cloning into pBKCMV plasmid vector. Expression
plasmids containing the complete protein-coding region of ER or
5ER were constructed by ligating the common N-terminal portion of
ER to the isoform-specific C-terminal portion via an overlapping
unique HindIII site. Positive clones were identified and
confirmed by dideoxy sequencing using primers covering the entire
translated region of ER. Reporter plasmid EREtk81Luc contains two
copies of consensus ERE palindromic sequence (aggtcacagtgacct) upstream
of the minimal thymidine kinase (tk) promoter in pA3Luc (courtesy of R.
Pestell, Albert Einstein college of Medicine, Bronx, NY). C3T1Luc of
the human C3 promoter (-1030 to +58) contains one copy of consensus
ERE palindromic sequences and two non-ERE E2 response
regions (courtesy of D. P. McDonnell, Duke University Medical
School, Durham, NC). The human c-fosLuc reporter contains
-2000 bp upstream of the transcriptional start site to 42 bp of human
c-fos promoter in pGLBasic (courtesy of C. Chen, University
of Queensland, Queensland, Australia).
In Vitro Translation of ER and 5ER
Variant
pBKCMV-ER or -5ER and the control pBKCMV (1 µg) were
used as DNA templates in coupled reticulocyte
lysate-transcription/translation reactions in the presence of T3 RNA
polymerase, amino acid mixture minus methionine, and
[35S]methionine in a final volume of 25 µl according to
the manufacturer protocols (Promega, Madison, WI). The synthesized
protein products were analyzed by fractionation on a 8.5% SDS-PAGE,
and the gels were dried down and autoradiographed for 24 h.
Stable Transfection and Screening
The expression vectors pBKCMV-ER and its 5ER variant
were stably transfected in ER/ERß-negative human osteosarcoma cell
line U2-OS (ATCC, Rockville, MD) to generate [ER]OS and
[5ER]OS cell lines. One x 106 OS cells were
washed with phenol red-free reduced serum OptiMEM and incubated with
the mixture of 30 µl lipofectamine (GIBCO BRL, Grand Island, NY) and
10 µg of pBKCMV-ER or its 5ER variant expression plasmid at
37 C for 24 h. Medium was replaced with serum-free phenol red-free
DMEM and incubated for an additional 48 h. Cells were subcultured
in a series of cell-plating density ranging from 5,000–50,000 cells
per 10-cm culture dish, in phenol red-free DMEM containing 5%
charcoal-treated FCS and geneticin G418 (GIBCO BRL) at 500 µg/ml for
selection. Media were then changed every 4–5 days. In 3 weeks, visible
colony foci were isolated and propagated in medium containing G418.
Immunocytochemistry
Parental U-2 OS cells and stably transfected [ER]OS and
[5ER] OS cells were grown on four-chamber cell culture slides
(Nunc Inc., Naperville, IL) and fixed with 2% paraformaldehyde in PBS
pH 7.2, for 24–48 h at 4 C. Before immunostaining, cells were washed
with PBS, pH 7.2, containing 10 mM glycine for 15 min to
quench unreacted aldehyde groups and rinsed twice with PBS. Cell
membranes were permeabilized by 5-min incubation at room temperature
with 0.2% Triton X-100 in PBS, and cells were washed twice with PBS
without detergent. Nonspecific binding sites for IgG were blocked by
incubating cells for 20 min with nonimmune serum (5%). Immunostaining
by avidin-biotin-peroxidase complex technique was performed according
to manufacturer’s instructions using Vectastatin Elite ABC
kit (Vector Laboratories, Burlingame. CA). The ER-monoclonal mouse
antibody, ER-1 (Babco, Richmond, CA) against the common N terminus
(amino acids 22–43 of the A/B domain)) of ER and 5ER, was
diluted to the optimal concentration in PBS. Cells were incubated with
the primary antibody for 1 h at 37 C and were washed twice with
PBS for 15 min each. The diluted biotinylated secondary antibody was
applied for 30 min at room temperature, and cells were washed twice
with PBS before a 30-min incubation with Vectastatin Elite ABC reagent.
Slides then were incubated in 1 µg/ml diaminobenzidine-0.3%
H2O2 for 5 min, rinsed, and counterstained with
hematoxylin. Coverslips were mounted to the chamber slides with
immumount (Shandon, Pittsburgh, PA). Immunoabsoption controls were
performed by preabsorption of the ER antibody ER1 overnight at 4 C
with the excess in vitro translated normal ER protein.
Method controls included substitution of nonimmune serum for primary
antibody, elimination of secondary antibody, or
streptavidin-biotin-peroxidase complexes, and dilution of primary
antibodies. All controls verified the specificity of the ER1 antibody
against the N terminus of ER and 5ER.
Transient Transfection and Luciferase Assay
[ER-]OS and [ER+]OS cells were plated in six-well
plates at a density of 2 x 105 cells per well and
allowed to adhere overnight. One hour before transfection, cells were
washed and incubated in phenol red-free reduced serum OptiMEM. A lipid
transfection mixture was prepared using a 1:400 mixture of
Dioleoyl-a-phosphatidylethanolamine to
demethyldioctadecylammonium bromide dissolved in 100% ethanol
(both lipids were purchased from Sigma Chemical Co., St. Louis, MO).
The DNA-lipid mix (1:5 ratio) containing the appropriate pBKCMV/ER
expression plasmid along with Luciferase reporter plasmid was prepared
in phenol red-free reduced serum OptiMEM (1 ml/well) for 30 min at room
temperature before addition to triplicate wells containing U2-OS cells.
Whenever applicable, empty pBKCMV plasmid was used for experiments
using pBKCMV-ER or its variant to ensure that stiochiometrically
equal amounts of CMV promoter, plasmid DNA, and lipid mix were applied
to each well. Rous sarcoma virus/ß-galactosidase was used to
monitor the variability of transfection efficiency. After 5 h
incubation at 37 C, the transfection medium was replaced with
serum-free phenol red-free DMEM and incubated for an additional 24
h, in the presence or absence of the indicated amount of
E2. Transfected cells were lysed with 300 µl lysis buffer
containing 1% Triton X-100, 10% glycerol, 2 mM EDTA, 2
mM dithiothreitol (DTT), and 25 mM
Tris-phosphate (pH 7.8), and the cellular debris was removed by
centrifugation. One hundred microliters of cell lysate were assayed for
luciferase activity by measuring light emission with Luminometer (EG&G
Berthold, Gaithersburg, MD) in the presence of luciferin and ATP.
EMSAs
Cell nuclear extracts were prepared from the parental U-2 OS,
stable [ER+]OS, and [5ER]OS cells grown in the absence or
presence of 10 nM E2 using the miniextraction
method with salt concentration modified for ER extracts (43).
Briefly, 5 x 106 to 1 x 107 cells
were collected, washed with ice-cold Tris-borate-saline, and
pelleted by centrifugation at 1,500 x g for 5 min. The
pellet was resuspended in 400 µl cold buffer A containing 10
mM HEPES, pH 7.9, 10 mM KCL, 0.1 mM
EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM
phenylmethylsulfonyl fluoride, and 50 µl protease inhibitor cocktail
(Sigma). Cells were placed on ice for 15 min, followed by addition of
25 µl 10% Nonidet NP-40, and were vortexed for 10 sec. After
centrifugation for 30 sec in a microfuge, the nuclear pellet was
resuspended in 50 µl ice-cold buffer C (20 mM HEPES, pH
7.9, 0.1 M NaCl, 1 mM EDTA, 1 mM
EGTA, 1 mM DTT, 1 mM phenylmethylsulfonyl
fluoride) and lysed for 15 min at 4 C on a shaking platform. The
nuclear extract was centrifuged for 5 min at 4 C, and the supernatant
was stored at -80 C. Protein concentrations of the nuclear extracts
were determined by Bradford microassay (Sigma).
Consensus and mutant ERE oligonucleotides were end-labeled with
[
32P]ATP and polynucleotide kinase (50,000 cpm/ng).
Mutant ERE was identical to consensus ERE oligonucleotide with the
exception of an AG-to-CC substitution in the ERE consensus sequence.
Labeled probe (10,000 cpm) was added to 20 µl reaction mixture
containing 1–2 µg nuclear extract in EMSA binding buffer (20
mM HEPES, pH 7.6, 400 mM KCL, 1 mM
DTT, and 20% glycerol) and 1 µg
poly(deoxyinosinic-deoxycytidylic)acid (Pharmacia Biotech,
Piscataway, NJ). Binding reactions were allowed to proceed at 25
C for 20 min. For competition studies, cold ERE was incubated with the
appropriate nuclear extract before addition of labeled ERE probe.
DNA-protein complexes were resolved by electrophoresis (150 V at 25 C)
through a nondenaturing polyacrylamide gel containing 5% glycerol in
0.5x Tris-borate-EDTA and were visualized by autoradiography.
For gel supershift analysis, 1–2 µl of appropriate gel supershift
antibodies were added per 20 µl reaction volume subsequent to
addition of 32P-labeled probe and incubated for 30 min at
25 C. Gel supershift antibodies included two monoclonal ER
supershift antibodies (Babco) against the N terminus (ER-2, directed
to amino acids 29–43 of the A/B domain) and the C terminus (ER-6,
directed against amino acids 575–595 of the EF region).
Cell Growth and Doubling Time Determination
Stably transfected [ER+]OS cells,
[5ER+[
ß]OS cells, and parental [ER-]Os cells
were seeded in 12-well plates at a density of 10,000 cells per well in
phenol red-free DMEM containing 5% charcoal-treated FCS and
antibiotics. Cells in triplicate wells were harvested every 2 days, and
nuclei were released with two drops of Zapoglobin per well (Coulter
Co., Miami, FL) and Coulter counted (Coulter Electronics, Hialeah, FL),
as previously described (44). The cell-doubling time (D),
obtained during log phase of the growth curve, was calculated
using a formula (45): D = 0.693/y (in hours), where y =
2.3 x (log ct1 - log ct0)/t1
- t0; ct1 = cell number at time 1;
ct0 = cell number at time 0; t1 = time (h) at
measurement of cells at time 1; t0 = time (h) at
measurement of cells at time 0.
|
FOOTNOTES |
|---|
This work was supported in part by an American Cancer Society, Massachusetts Division, Research Grant Award (to J.M.A.), The Jarislowsky Foundation, and a Massachusetts General Hospital Medical Discovery Award (to S.S.C.).
Portions of this work were presented at the 80th Annual Meeting of The Endocrine Society, Minneapolis MN, June 1997.
Received for publication March 6, 1998. Revision received May 7, 1998. Accepted for publication May 26, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. P. Gillam, M. E. Molitch, G. Lombardi, and A. Colao Advances in the Treatment of Prolactinomas Endocr. Rev., August 1, 2006; 27(5): 485 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Han, L. Ding, B. Yuan, X. Yang, X. Wang, J. Li, Q. Lu, C. Huang, and Q. Ye Hepatitis B virus X protein and the estrogen receptor variant lacking exon 5 inhibit estrogen receptor signaling in hepatoma cells Nucleic Acids Res., June 6, 2006; 34(10): 3095 - 3106. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. C. Peterson, M. D. Servinsky, A. Christian, Z. Peng, W. Qiu, J. Mann, J. Dicello, and D. L. Huso Tamoxifen resistance and Her2/neu expression in an aged, irradiated rat breast carcinoma model Carcinogenesis, September 1, 2005; 26(9): 1542 - 1552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Bryant, A. E. Snowhite, L. W. Rice, and M. A. Shupnik The Estrogen Receptor (ER){alpha} Variant {Delta}5 Exhibits Dominant Positive Activity on ER-Regulated Promoters in Endometrial Carcinoma Cells Endocrinology, February 1, 2005; 146(2): 751 - 759. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Widder, T. Pelzer, C. von Poser-Klein, K. Hu, V. Jazbutyte, K.-H. Fritzemeier, C. Hegele-Hartung, L. Neyses, and J. Bauersachs Improvement of Endothelial Dysfunction by Selective Estrogen Receptor-{alpha} Stimulation in Ovariectomized SHR Hypertension, November 1, 2003; 42(5): 991 - 996. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kiaris, I. Chatzistamou, A. V. Schally, G. Halmos, J. L. Varga, H. Koutselini, and A. Kalofoutis Ligand-dependent and -independent effects of splice variant 1 of growth hormone-releasing hormone receptor PNAS, August 5, 2003; 100(16): 9512 - 9517. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Garcia Pedrero, P. Zuazua, C. Martinez-Campa, P. S. Lazo, and S. Ramos The Naturally Occurring Variant of Estrogen Receptor (ER) ER{Delta}E7 Suppresses Estrogen-Dependent Transcriptional Activation by Both Wild-Type ER{alpha} and ER{beta} Endocrinology, July 1, 2003; 144(7): 2967 - 2976. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. O. Mueller, J. M. Hall, D. L. Swope, L. C. Pedersen, and K. S. Korach Molecular Determinants of the Stereoselectivity of Agonist Activity of Estrogen Receptors (ER) alpha and beta J. Biol. Chem., March 28, 2003; 278(14): 12255 - 12262. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Yang, A. Kriatchko, and S. K. Roy Expression of ER-{alpha} and ER-{beta} in the Hamster Ovary: Differential Regulation by Gonadotropins and Ovarian Steroid Hormones Endocrinology, June 1, 2002; 143(6): 2385 - 2398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Campbell-Thompson, I. J. Lynch, and B. Bhardwaj Expression of Estrogen Receptor (ER) Subtypes and ER{beta} Isoforms in Colon Cancer Cancer Res., January 1, 2001; 61(2): 632 - 640. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Bollig and R. J. Miksicek An Estrogen Receptor-{alpha} Splicing Variant Mediates Both Positive and Negative Effects on Gene Transcription Mol. Endocrinol., May 1, 2000; 14(5): 634 - 649. [Abstract] [Full Text]
|
|
|
|
 |
|
 D. C. Danila, X. Zhang, Y. Zhou, G. R. Dickersin, J. A. Fletcher, E. T. Hedley-Whyte, M. K. Selig, S. R. Johnson, and A. Klibanski A Human Pituitary Tumor-Derived Folliculostellate Cell Line J. Clin. Endocrinol. Metab., March 1, 2000; 85(3): 1180 - 1187. [Abstract] [Full Text]
|
|
|
|
|
|
S. C. Sharma, J. W. Clemens, M. D. Pisarska, and J. S. Richards Expression and Function of Estrogen Receptor Subtypes in Granulosa Cells: Regulation by Estradiol and Forskolin Endocrinology, September 1, 1999; 140(9): 4320 - 4334. [Abstract] [Full Text]
|
|
|
|
|
|
B. Carlson, T. Lahusen, S. Singh, A. Loaiza-Perez, P. J. Worland, R. Pestell, C. Albanese, E. A. Sausville, and A. M. Senderowicz Down-Regulation of Cyclin D1 by Transcriptional Repression in MCF-7 Human Breast Carcinoma Cells Induced by Flavopiridol Cancer Res., September 1, 1999; 59(18): 4634 - 4641. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Balleine, S. M. N. Hunt, and C. L. Clarke Coexpression of Alternatively Spliced Estrogen and Progesterone Receptor Transcripts in Human Breast Cancer J. Clin. Endocrinol. Metab., April 1, 1999; 84(4): 1370 - 1377. [Abstract] [Full Text]
|
|
|
|
|
|
D. A. Schreihofer, E. M. Resnick, A. Y. Soh, and M. A. Shupnik Transcriptional Regulation by a Naturally Occurring Truncated Rat Estrogen Receptor (ER), Truncated ER Product-1 (TERP-1) Mol. Endocrinol., February 1, 1999; 13(2): 320 - 329. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
