| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Medical Nutrition (S.S., H.S., S.A., J.-Å.G.), Department of Bioscience (M.W., J.-Å.G.), Karolinska Institute, Novum, S141–86 Huddinge, Sweden
Address all correspondence and requests for reprints to: Dr. Jan-Åke Gustafsson, Department of Medical Nutrition, Karolinska Institute, F60, Novum, S141–86 Huddinge, Sweden. E-mail: jan-ake.gustafsson{at}mednut.ki.se
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ß proteins (ER and ERß) at various
stages of development of the rat mammary gland were quantified by
Western blotting. ER and ERß recombinant proteins were used as
standards, and their molar concentrations were measured by ligand
binding assays. In 3-week-old pregnant, lactating, and postlactating
rats the ER content ranged from 0.30–1.55 fmol/µg total protein
(mean values). The ERß content of the same samples ranged between
1.06–7.50 fmol/µg total protein. At every developmental stage, the
ERß content of the mammary gland was higher than that of ER. When
receptor levels were normalized against ß-actin, it was evident that
ER expression changed during development, with maximum expression of
both receptors during the lactation period. With an antibody raised
against the 18-amino acid insert of the ERß variant, originally
called ERß2 but named ERßins in this paper, Western blots revealed
that ERßins protein was up-regulated during the lactation period.
RT-PCR showed that the levels of messenger RNA of ERßins paralleled
those of the protein. Double immunohistochemical staining with
anti-ER and anti-ERßins antibodies revealed that ERßins protein
colocalized with ER in 70–80% of the ER-expressing epithelial
cells during lactation and with 30% of these cells during pregnancy.
These observations indicate that expression of ERßins is regulated
not only quantitatively, but also with regard to its cellular
distribution. As ERßins acts as the dominant repressor of ER, we
suggest that its coexpression with ER quenches ER function and
may be one of the factors that contribute to the previously described
insensitivity of the mammary gland to estrogens during lactation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ERß, cloned in 1986 (3) and
1996 (4), respectively. In their DNA-binding domains
(5), ER and ERß share 96% homology and recognize the
same estrogen-responsive element (ERE) on genomic DNA. In their
ligand-binding domains (LBD) there is only 60% homology, and this
results in some degree of ligand selectivity between the receptors;
indeed, compared with ER, ERß has a lower affinity for estradiol
(E2), but a higher affinity for genistein
(5, 6). For full transcriptional activity in an ERE
reporter assay using 293 cells with transfected ERs, a 10-fold higher
concentration of E2 is required for ERß than
for ER, whereas a 10-fold lower concentration of genistein can
activate ERß (7). Although, on classical EREs, both
ER and ERß activate transcription, they can work in opposite
directions on activating protein-1 response elements (8).
In the presence of E2, ER is an activator, but
ERß is an inhibitor or silencer, at activating protein-1 sites
(8). Moreover, recent publications provide evidence that
ERß is not simply an ER that utilizes ligands other than those used
by ER, but is an independent regulator with distinct cellular
functions (9, 10, 11). There are three potential translational start sites at the 5'-end of rat ERß messenger RNA (mRNA), which, if used, would produce proteins 485, 530, and 549 amino acids (aa) in length, differing at their N-termini (12). Similar sites are found in the human ERß sequence and result in polypeptide chains that are 477, 485, and 530 aa in length (12, 13). Fuqua et al. reported that 530- and 477-aa proteins can be produced in cells transfected with the human full-length complementary DNA (cDNA) (14). However, it is not yet clear whether products of different sizes are normally expressed in tissues. From the few studies in which Western blotting of tissue extracts have been made, the longest ERß form seems to be the major component (14, 15). As differences at the N-terminus may affect the ligand-independent transcriptional activity of the activating factor-1 domain (16), this issue needs further analysis.
Although these N-terminal differences do not change the ligand-binding
character of ERß, some of the alternatively spliced variants, which
have been reported in human, mouse, and rat, have weak affinities for
E2 and seem to be very important in terms of
their relation to the function of ER. One ERß variant, first
reported as ERß2 by Petersen et al. (17) and
named ERßins in this paper, has a 54-bp (18-aa) additional in-frame
sequence between exons 5 and 6 of wild-type (wt) rat ERß. This
insertion causes about 1/35-fold lower binding affinity for
E2 (17, 18). As the capacity for DNA
binding and dimerization with ER and wt ERß remains intact,
ERßins can act as a negative regulator when it is coexpressed with
ER or wt ERß (17, 18). In humans, a C-terminally
truncated variant, named ERß2 by Moore et al.
(19) and ERßcx by Ogawa et al.
(20), does not bind E2 and is a
dominant negative regulator for ER, but not for wt ERß (19, 20).
Information about the tissue distribution of ER and ERß proteins
is necessary for understanding the specific physiological functions of
these two receptors. Granulosa cells of ovary and epithelial cells of
ventral prostate express ERß exclusively (4, 21), and
analysis of these tissues can provide information about which genes are
ERß regulated. On the other hand, in the mammary gland, at least in
humans and rats, both ER and ERß are expressed (15, 22, 23, 24). In this case, interaction between the two receptors in
individual cells may be very important for the overall effect of
estrogen on the whole tissue.
Recent reports suggested the importance of wt ERß as a regulator of
ER function. For example, Pettersson et al. reported that
wt ERß can dose dependently down-regulate ER’s
trans-activation potency when the concentration of
E2 is not sufficient for activation of ERß
(7). As described previously, rat ERßins and human
ERßcx are dominant negative regulators of ER independent of the
E2 concentration (17, 18, 19, 20). All of
these findings indicate that for an understanding of the role of ERs in
tissues where ER and ERß are coexpressed, information is needed
about the degree of colocalization of ER with ERß and its variants
and on the relative quantities of the two receptors.
In this study we evaluated the expression and colocalization of ER,
ERß, and one ERß variant, ERßins, in rat mammary gland using
quantitative Western blotting and immunohistochemistry. The results
suggest a possible role for ERßins in silencing of ER function
during the lactating period.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant proteins as positive control
Recombinant ER and ERß1 (ERß 530) proteins were purchased
from Panvera (Madison, WI). The amount of functional receptor was
measured by a tritium-labeled E2 binding assay.
Both of the human recombinant proteins were made from
baculovirus-infected cells and had ER concentrations of 7.3 pmol/µl
(ER) and 1 pmol/µl (ERß 530). The ERß 503 protein is composed
of the human 485-aa sequence with the rat 18 aa of rat ERßins
(17). This was produced by overexpression in SF9 cells by
Karo Bio AB (Huddinge, Sweden). This form of ERß (ERßins) has very
little binding affinity for E2, but its affinity
for 4-hydroxytamoxifen is similar to that of ERß 530. All proteins
were divided into small aliquots and thawed only once to avoid
degradation.
Antibodies
ER 6F11, a mouse monoclonal antibody, was obtained from Novo
Castra (Newcastle upon Tyne, UK). ERß LBD rabbit polyclonal antibody
was produced by us as described previously (15). To remove
any cross-reactivity of ERß LBD antibodies with ER protein, 1 µl
LBD antibody was preincubated with 10 pmol ER recombinant protein
for 8 h at 4 C before use.
ERß INS chicken polyclonal antibody was raised against the rat 18-aa insert sequence of ERßins (17). Mouse monoclonal ß-actin antibody AC-15 and horseradish peroxidase-conjugated secondary antibodies were purchased from Sigma (St. Louis, MO). Fluorescence-labeled secondary antibodies for double staining were obtained from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA).
Protein extraction for Western blotting
For efficient extraction of all nuclear receptors from the
tissue and for the visualization of sharp bands that can be correctly
measured by the photoimager, modifications were made to the protocols
previously reported (15). Tissues were removed from
animals, placed in ice-cold lysis buffer, and homogenized immediately.
Lysis buffer was composed of 600 mM Tris-HCl, 1
mM EDTA (pH 7.4), and two protease inhibitor tablets
(Roche Molecular Biochemicals, Mannheim, Germany)/50 ml.
The homogenates were centrifuged for 1 h at 105,000 x
g, and the protein concentrations of supernatant were
measured by protein assay (Bio-Rad Laboratories, Inc.,
Hercules, CA) with BSA as standard.
Direct usage of this sample on SDS-PAGE resulted in less stacking as well as fuzzy protein bands, probably due to the high salt concentration. To overcome these problems, 65 µg of each sample were precipitated with trichloroacetic acid (10% final) and resuspended in 100% methanol. Samples were placed on dry ice for 30 min, and the protein was recovered by centrifugation. Precipitates were dissolved in 1 x SDS sample buffer and used for SDS-PAGE.
Western blotting analysis
Samples were resolved on SDS-polyacrylamide gels, either 9%
polyacrylamide or preformed gradient gels 4–20% acrylamide
(Novex, San Diego, CA), with a Tris-glycine buffer system.
Transfer to ProBlot membranes (PE Applied Biosystems,
Foster City, CA) by electroblotting was performed either by semidry or
wet blotting or in a Tris-glycine buffer. Kaleidoscope prestained
standards (Bio-Rad Laboratories, Inc.) and a low mol wt
electrophoresis calibration kit (Amersham Pharmacia Biotech, Uppsala, Sweden) were used as mol wt markers.
After 1-h blocking with blocking buffer (10% skim milk in PBS with
0.1% Nonidet P-40) at room temperature, membranes were incubated with
a 1:75 dilution of ER 6F11 mouse antibodies, a 1:3000 dilution of
ERß LBD rabbit antibodies, or a 1:500 dilution of ERßINS chicken
antibodies in blocking buffer at 4 C overnight. This was followed by
1-h washing in blocking buffer and then incubation in a 1:7500 dilution
of horseradish peroxidase-conjugated secondary antibodies for each
species in blocking buffer for 1 h at room temperature. After
sequential washing with blocking buffer, PBS with 0.1% Nonidet P-40,
and PBS alone, signals were developed using ECL Plus (Amersham Pharmacia Biotech).
For normalization against an intracellular protein, the membranes used
for ER or ERß detection were sequentially probed with a
ß-actin-specific antibody without stripping the membrane as described
by Liao et al. (25). Bands on the exposed film
were captured with an image analyzer (LAS-1000, Fuji Photo Film Co. Ltd., Tokyo, Japan), and densitometric values were measured
with Image Garge (Fuji Photo Film Co., Ltd.). Three sets
of densitometric values obtained from three separate blots were used
for the calculation.
Calculation of protein amount from Western blotting results
By referring to the data and graph obtained from the control
study (left graph in Fig. 1B
)
and the densitometric values of the recombinant protein on the blots
(Fig. 2, A and B), an approximate formula
was drawn by polynomial regression with the StatView program (Abacus
Concepts, Inc., Beverly, CA). From triplicate experiments using
recombinant proteins, the formula that had the highest
r2 value was chosen (r2 =
0.996; P < 0.01 for ER and r2
= 0.997; P < 0.01 for ERß). The molar concentration
of ER in each lane was calculated by the selected formula and expressed
as femtomoles per µg total protein extract. From the calculated value
of three independent Western blots using three sets of samples, the
mean and SD were calculated.
|
|
. Statistical analysis was performed by unpaired
Welch’s t test.
|
The primer sets for detecting both wt and ßins variants in one reaction were rERß LBD U (5'-GAGCTCAGCCTGTTCGACC) and rERß LBD L (5'-GGCCTTCACACAGAGATACTCC) (18). The primer sets for detecting only ERßins were rERß INS U (5'-ATTATAAATATACTATATCATTAATATTAATCCTCAGAAGACCCT) and rERß INS L (5'-ATATTA ATATATGTATATTAATTATTTAATGGGCAGCACTCTTCA). For detection of the total amount of ERß mRNA, we used rERß EX6 U (5'-AATCTTTGACATGCTCCTGGCG) and rERß EX7 L (5'-AAAGAAGCATCAGGAGGTTGGC). The sizes of respective products are 246 bp for wt and 300 bp for ßins when using rERß LBD U and L primers. Combination of rERß INS U and L primers produces the 110-bp product, and rERß EX6 U with EX7 L primer makes the 263-bp band as a total ERß. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) primers were 5'-CAGCCGCATCTTCTTGTG for sense and 5'-AGTTGTCATATTTCTCGTGGTTCA for antisense.
PCR reactions were 95 C for 5 min, cycle step, and 72 C for 5 min with TaqStart antibody for hot start (CLONTECH Laboratories, Inc. Palo Alto, CA). Cycle steps for each primers were as follows: rERß LBD U and L, 36 cycles at 95 C for 30 sec, 57 C for 40 sed, and 72 C for 60 sec; rERß INS U and L, 39 cycles at 95 C for 30 sec, 51 C for 30 sec, and 72 C for 50 sec; rERß EX6 and EX7U, 35 cycles at 95 C for 30 sec, 57 C for 30 sec, and 72 C for 60 sec; and GAPDH, 26 cycles at 95 C for 30 sec, 56 C for 40 sec, and 72 C for 60 sec. Cycle numbers were decided by the control study, which confirmed the linear phase of reaction (data not shown). For reactions with rERß INS U and L primers, a minimum of 38 cycles was needed for detection of the band.
PCR products were resolved on a 2% agarose gel containing ethidium
bromide in 0.5 x Tris-borate EDTA running buffer. Products
as total ERß and ERßins were loaded in one lane on an agarose gel
(second panel of Fig. 5B). The 123-bp DNA ladder obtained
from Life Technologies, Inc. was used for calibration. The
negative control in each set was a reaction where the cDNA template was
replaced with DNase-treated RNA (data not shown), and the identity of
the positive band was confirmed by direct sequencing of the
representative PCR product or by cloning the PCR product into a TA
cloning vector, selection of clones, and sequencing of the insert from
the vector.
|
Immunohistochemical staining
Double immunohistochemical staining using
fluorescence-conjugated secondary antibody was performed as described
previously (15). Briefly, frozen sections from rat mammary
gland were fixed with ice-cold methanol, acetone, and 4%
paraformaldehyde. After treatment with 0.5% Triton X-100 in PBS and
10% normal serum from the same species as the secondary antibody,
sections were incubated sequentially as follows: ER 6F11 mouse
antibody (1:100), fluorescein isothiocyanate (FITC)-conjugated
antimouse IgG (1:50), ERß INS chicken antibody (1:500), and
Cy3-conjugated anti chicken IgY (1:200). Nuclear staining was performed
with 0.1 µg/ml 4',6-diamidino-2-phenylindole (DAPI). The sections
were examined under the fluorescence microscope with suitable filters
for FITC, Cy3, and DAPI, and images captured by CCD camera were
analyzed with OpenLab 2.03 program (Improvision, Covently, UK). To have
accurate evaluation, counting of positively labeled cells was assisted
by this program, which could clearly mark the labeled cells less
ambiguously than the human eye. Counting analysis was performed on six
pictures produced from three individual samples in each category.
For the control study of our antibody, 1 µl ERß INS antibody was incubated with an excess of ERß 503 protein or ERß 530 protein in PBS at 4 C overnight. These preadsorbed antibodies with PBS were applied to sections at 1:200–300 dilution.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and ERß
proteins expressed in tissues are available to date. Such information
is essential for evaluation of functional interaction between these
receptors. To determine their relative abundance in rat mammary gland,
we developed a quantitative Western blot assay for ER and ERß. By
comparison of the band densities of recombinant proteins whose molar
concentrations were predetermined by binding assays and those of
samples, we could calculate the content of receptor protein in tissue
samples. The receptor concentrations in recombinant human ER and
ERß proteins, purchased from Panvera, were assessed by a
tritium-labeled E2 binding assay. As shown in
Fig. 1A
, both recombinant proteins showed a clear single band with no
evidence of degradation. The relationship between the densitometric
value of ER and ERß bands on the blots and the amount of protein
is shown in Fig. 1B. From these results, a regression line was drawn,
and a formula for calculating the amount of receptor protein from the
band densities was made (left graph in Fig. 1B). The
specific binding activity of Panvera ERß is 6500 pmol/mg protein.
With a molecular mass of 63 kDa, there are approximately 16,000 pmol
ERß protein for each 6500 pmol binding activity. This means that 40%
of the ERß is functional binding protein. For each picomole of
binding protein loaded onto the lane, 2.5 pmol ERß protein are
loaded. We are, therefore, underestimating our tissue content of ERß
by a factor of 2.5. For Panvera ER the binding activity is 12,300
pmol/mg protein. It can be calculated that there are 15,000 pmol ER
for each 12,300 pmol of binding activity. This means that 82% of ER
is functional binding protein. For each picomole of binding protein,
loaded onto a lane, we are actually loading 1.25 pmol ER protein. We
are, therefore, also underestimating the amount of ER in the tissue,
but not by much.
Evaluation of ER and ERß proteins by quantitative Western
blotting
Aliquots containing 65 µg total protein from extracts of mammary
glands at each developmental stage were loaded onto gels with
calculated amounts of recombinant protein in adjacent lanes (Fig. 2).
On Western blots, the ER antibody detected a single band in tissue
samples, and this band comigrated with recombinant ER protein, which
migrated as a 66- to 67-kDa band (Fig. 2A). With the ERß LBD
antibody (Fig. 2B) there were three bands in rat mammary gland tissues.
Figure 2D is a high magnification of the bands of the blot of the three
individual samples from 3-week-old rats. The main bands were in the
62-kDa range. The largest band was very weak, with a size of
approximately 65 kDa. As shown in Fig. 2B, ERß LBD antibody did not
recognize 0.5 pmol ER protein, but clearly detected 0.5 pmol ERß
530 protein, which migrated as a 62-kDa band. The 65-kDa band,
therefore, also due to the difference in size, cannot be explained as a
cross-reaction with ER. As the most likely explanation for this band
is that it represented nonspecific reaction, we did not include it in
the quantitative analysis. ERß LBD antibody also detected recombinant
ERß 503 protein, which is composed of the 485-aa sequence of human
ERß with an additional 18-aa rat sequence of ERßins in LBD (Fig. 3). ERß 503 protein was observed as a
band around 58–59 kDa. This indicates that ERß LBD polyclonal
antibody must recognize not only wt, but also ERßins protein. In
addition, by comparison of the signal of ERß 530 with that of the LBD
antibody, we could make the rough estimate that the amount ERß 503
loaded per lane was approximately 0.5 pmol. From these findings, we
conclude that the main 62-kDa bands probably represent 1) wild-type
ERß, translated from the first methionine of mRNA resulting in a
530-aa-long protein; and 2) ERßins, expected to have 530 plus 18
aa. We observed that a difference of 18 aa between two ERß
proteins is not sufficient to allow separation by SDS-PAGE on 9% gels
(data not shown). In addition, we developed a specific antibody that
only detects ERßins protein. ERß INS antibody was raised against
the specific 18-aa polypeptide of rat ERßins (17). This
antibody detected ERß 503 protein, but not ERß 530 protein (Fig. 2C). In mammary gland samples, this antibody recognized two bands, one
around 62 kDa and the other one around 58–59 kDa (Fig. 2C). The
relative intensities of these two bands varied between samples.
Although the 62-kDa band migrates with the main band observed with the
LBD antibody and is expected to be 530 plus 18 aa, the 58- to 59-kDa
band is only observed with ERß INS antibody. The source of this lower
band is uncertain, and it is being further investigated. It could be an
ERßins composed of 485 plus 18 aa (503 aa in total), a combination of
an exon 3 deletion with ERßins (17), or a degraded form
of ERßins that could not be detected by LBD antibody.
|
and 0.10–0.50 pmol/lane for ERß. As
is evident from the results shown in Table 1, ERß was more abundant than ER at
every developmental stage of the mammary gland.
|
, when equal amounts of cytosolic protein
from mammary glands at various stages were loaded in each lane, the
increase in the receptor content of the gland at lactation was not
evident. This is because at lactation milk proteins contribute
significantly to the soluble proteins, and the amount of truly
cytosolic protein loaded in the samples is less than that at other
stages. In an attempt to circumvent the problem of milk protein
contamination of the cytosol, all densitometric values were normalized
against ß-actin. The ß-actin level in each sample was evaluated by
Western blotting of the membranes after analysis of ER or ERß.
With ß-actin as an internal standard it is clear that both ER and
ERß proteins had a peak of expression during lactation (Fig. 4, A and B). With the same
standardization procedure, Western blots with ERßins-specific
antibody revealed that the expression of ERßins protein seemed to be
regulated independently of wt ERß (Fig. 4C).
Changes in mRNA expression
To determine whether transcription was involved in the change in
expression of wt ERß and ERßins, an RT-PCR study detecting wt ERß
and ERßins mRNAs was performed. The first set of PCR primers is
designed for detecting both wt and ßins in one reaction and for
evaluating the ratio of these mRNAs (Fig. 5A) (18). However, although
Lu et al. reported the exact expression ratio of these mRNAs
by using essentially the same approach for mouse samples
(26), our experience has been that the ratios of wt to
ßins were not reproducible when the difference in amounts of wt and
ßins were small. Accordingly, we present these data to indicate the
relative abundance of wt and ßins mRNAs. As shown in the upper
panel of Fig. 5B, the balance between wt and ßins expression
changed during mammary gland development. It is notable that 3-week-old
(immature) rats as well as rats during early pregnancy and late
lactation show increased relative levels of ERßins mRNA, whereas 4-
and 6-week-old rats seem to have a low level of ERßins mRNA. To
confirm this tendency, we developed two other sets of primers. One set
detects the total amount of ERß, and the second set detects only the
insert sequence of ERßins. PCR reactions with each set of primers
were performed independently, and the resulting products were loaded in
one lane on an agarose gel (second panel of Fig. 5B). As the
source of product amplified by ßins-specific primer is a very low
abundance message, the intensity of the band is not reliable as a
semiquantitative measurement. The lower PCR products shown in the
second panel of Fig. 5B is the result of 39-cycle
amplification using ßins-specific primers. An increase in cycle
number could not enhance the intensity of the bands in a linear way
(data not shown). Due to these difficulties of accurate quantification
of ßins mRNA, the tendency for a change in ERßins mRNA is expressed
diagrammatically at the bottom of Fig. 5C. Densitometric
values of the bands from total ERß primers were normalized against
those of GAPDH. Figure 5C shows the change in total ERß mRNA
expression. In the graph the mean value of 3-week-old rat samples is
arbitrarily set at 1. By comparison between the graphs of Fig. 4, B and
C, as well as Fig. 5C, it is apparent that although the variations in
levels of ERßins protein seem to be related to those of ERßins
mRNA, the expression of wt ERß protein is not directly correlated to
the levels of its mRNA. This suggests that mechanisms other than
transcriptional control, such as control of degradation and/or of
translational efficiency, are involved in regulation of wt ERß
protein.
Colocalization of ERßins with ER
In cellular systems, ERßins heterodimerizes with and is a
dominant negative regulator of ER and wt ERß (17, 18).To gain some insight into the possible physiological role of
ERßins, we performed double immunostaining using fluorescence-labeled
secondary antibodies to evaluate the colocalization of ERßins with
ER. Figure 6 is a control study with
lactating mammary gland showing the specificity of ERß INS antibody.
ERß INS antibody gave mainly nuclear staining, but there was, in
addition, some cytoplasmic staining that appeared as small dots. The
staining pattern was not changed by preadsorption of the antibody with
ERß 530 protein. However, all nuclear staining was abolished when the
antibody was preadsorbed with ERß 503 protein, which has the 18-aa
sequence of ERßins (Fig. 3). Some cytoplasmic staining was left after
preadsorption with ERß 503. These results suggest that the nuclear
staining is specific for ERßins. Figure 7 is a representative picture of double
staining using ER 6F11 mouse monoclonal antibody with ERß INS
chicken polyclonal antibody. The green, red, and yellow signals
indicate cells expressing ER alone, ERßins alone, and both
receptors, respectively. Not only epithelial cells, but also some
populations of stromal cells (adipocytes), were stained with ERß INS
antibody. To focus on the possible role of ERßins in ER function,
we calculated what percentage of ER-expressing epithelial cells also
expressed ERßins. Calculations were made from nuclei in six pictures
from three individual samples in each category (bottom of
Fig. 7). In the pregnant mammary gland only about 30% of
ER-containing cells express ERßins, indicating that in 70% of
ER-containing cells this receptor is free from ERßins. On the
other hand, during lactation, 70–80% of ER-containing
epithelial cells coexpress ERßins. This indicates that a major
part of ER in lactation may be antagonized by ERßins expressed in
the same nuclei, and this may contribute to the insensitivity of the
lactating mammary gland to estradiol that was reported 2 decades ago
(27, 28, 29).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,
the level of ERß mRNA in the mammary gland of mice and rats is low
(5, 17, 30). We also observed with PCR assays that the
level of ERß mRNA appears to be lower than that of ER. These PCR
studies showed that with ERß primers, at least 33 cycles of
amplification are needed for the detection, but with ER-specific
primers, PCR products were visible after 27 cycles (data not shown).
Upon evaluation of the receptor proteins, however, ERß appears to be
expressed in more cells than ER in rats and mice (15, 31). In the case of human samples, Speirs et al.
reported that ERß mRNA, evaluated by RT-PCR, and ERß protein,
analyzed by immunohistochemistry, were more frequently detected in
epithelial cells of normal mammary gland than ER
(23).
In this report we show that ERß protein is more abundantly expressed
than ER protein in the rat mammary gland. This is true before
puberty, after puberty, during pregnancy, during lactation, and during
the postlactation period. The methods used here are based on the
assumption that as proteins separated by SDS-PAGE have no
three-dimensional structure, antibodies recognize epitopes of
target protein in samples and in recombinant proteins with the same
efficiency. For ER detection we used ER 6F11, which was raised
against full-length human ER. We cannot rule out the possibility
that this antibody recognizes rat ER less efficiently than human
ER protein. However, with the homology between human and rat ER
of 88% (32) it is unlikely that with the entire protein
as antigen there would be significant differences in affinity of the
antibody for rat and human ER, respectively. Moreover, with another
ER antibody, ER MC-20 (Santa Cruz Biotechnology, Inc.), whose antigen is the C-terminal part of mouse ER, we
obtained results indistinguishable from those presented here (data not
shown). For ERß detection the same logic applies, as ERß LBD
antibody was raised against human ERß LBD, whose homology to rat is
93% (5). Even if it may be argued that the numbers
generated by our assays do not constitute an accurate representation of
the cellular concentrations of the two ERs, it is nonetheless clear
that ERß protein is more abundantly expressed than ER protein. The
differences range from a minimum of 2-fold to a maximum of 10-fold.
In an earlier report we showed Western blotting of ERß using similar samples; with that protocol for sample preparation, however, we could not evaluate the molecular size and accurately measure variations in ERß protein (15). When we use a high salt lysis buffer that effectively extracts all DNA-binding proteins, there is a problem of stacking and of broad fuzzy bands on SDS-PAGE (15). In the present study we overcame this problem with trichloroacetic acid, followed by methanol precipitation (33) and obtained bands that were sharp and strong enough for quantitative evaluation.
It is notable that the changes in ERß protein at various stages of mammary gland development were not always paralleled by changes in ERß mRNA. For example, although ERß mRNA increases in late lactation, ERß protein decreases. This discrepancy between mRNA and protein suggests the existence of other critical regulatory mechanisms, such as control of protein degradation and changes in efficiency of translation of mRNA.
Our working hypothesis is that one of the physiological roles of ERß
is to act as a negative regulatory partner of ER. We have already
presented evidence for this idea in the rat uterus (34).
In this context, ERß variants such as ERßins and ERßcx may be
more important than wt ERß, because they can heterodimerize with and
act as dominant negative regulators of ER (17, 18, 19, 20). To
be a negative regulator two conditions must be met. These are 1)
colocalization with ER in same nuclei, and 2) presence of at least
an equimolar ratio of ERß and ER proteins. Our double
immunostaining shows that the first condition is met in the lactating
mammary gland. In the pregnant mammary gland where ER induces
progesterone receptor (35, 36, 37), about 70% of ER is
expressed alone and is free from potential inhibition by ERßins.
However, during the lactation period, colocalization of ERßins with
ER is increased, and in 70–80% of ER-containing cells ER may
be antagonized by ERßins.
The second condition, an equimolar ratio of ER and ERßins, is more
difficult to demonstrate directly. However, as the molar ratio of ERß
to ER is approximately 3 during lactation, and there is a
significant increase in ERßins protein in the lactating period, it
seems clear that there is sufficient ERßins in ER-expressing
epithelial cells to affect ER function.
It is a well known observation that experimentally the lactating
mammary gland is estrogen insensitive, i.e. there is no
induction of progesterone receptor by doses of estradiol that are
effective in the mammary glands of age-matched virgins
(27, 28, 29). We present the hypothesis that antagonism of
ER by ERßins may be the source of this insensitivity.
An important question that arises from this hypothesis is whether there
are ligands that activate estrogen receptors during the period of
lactation. E2 levels are very low during
lactation (38), but there are several possible conditions
during which ERßins may be needed to prevent activation of ER.
These include high levels of phyto- or xenoestrogens (39)
and activation of ER via phosphorylation by epidermal growth factor- or
insulin-like growth factor-stimulated pathways (40, 41).
A recent paper reported that in the presence of
E2 or raloxifene, ERßins can activate
the transforming growth factor-ß promoter as efficiently as ER and
wt ERß (26). This indicates that in addition to
antagonism of ER, ERßins may have other functions.
In human tissues ERßins is rarely expressed, although ERßcx is
frequently observed in normal human breast tissue (24, 42). We speculate that human ERßcx has a role similar to that
we have suggested for rat ERßins. If in human lactating breast
ERßcx is also induced, antagonism of ER by ERßcx may occur. This
may contribute to the reduced risk of breast cancer in women who
breastfeed their babies, which has been reported in epidemiological
studies (43). Further investigation of ERß variants in
the normal human breast is clearly needed.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Recipient of a fellowship from Wenner-Gren Foundation
(Sweden) and a research grant from Scandinavia Japan Sasakawa
Foundation.
Received December 18, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
or ß.
Endocrinology 141:1839–1845 and estrogen
receptor-ß: correlations with biological character and distinct
differences among SRC coactivator family members. Endocrinology 141:3534–3545
in vivo and in vitro. Biochem Biophys Res Commun 243:122–126[CrossRef][Medline]
and ß in the rodent
mammary gland. Proc Natl Acad Sci USA 97:337–342 and ß. Mol Pharmacol 58:584–590 in the rat ovary.
Endocrinology 140:963–971 and PR and associated
with nodal status, grade, and proliferation rate in breast cancer.
Am J Pathol 156:29–35 and ERß status. J Histochem
Cytochem 47:1323–1330This article has been cited by other articles:
 |
|
 |
|
  T. F.G Lucas, E. R Siu, C. A Esteves, H. P Monteiro, C. A Oliveira, C. S Porto, and M. F. M Lazari 17Beta-Estradiol Induces the Translocation of the Estrogen Receptors ESR1 and ESR2 to the Cell Membrane, MAPK3/1 Phosphorylation and Proliferation of Cultured Immature Rat Sertoli Cells Biol Reprod, January 1, 2008; 78(1): 101 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. E. Connor, M. J. Meyer, R. W. Li, M. E. Van Amburgh, Y. R. Boisclair, and A. V. Capuco Regulation of Gene Expression in the Bovine Mammary Gland by Ovarian Steroids J Dairy Sci, June 1, 2007; 90(13_suppl): E55 - E65. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. N. Sundar, V. Kerekatte, C. N. Equinozio, V. B. Doan, L. F. Bjeldanes, and G. L. Firestone Indole-3-Carbinol Selectively Uncouples Expression and Activity of Estrogen Receptor Subtypes in Human Breast Cancer Cells Mol. Endocrinol., December 1, 2006; 20(12): 3070 - 3082. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Lopez, N. Torres, V. Ortiz, G. Aleman, R. Hernandez-Pando, and A. R. Tovar Characterization and regulation of the gene expression of amino acid transport system A (SNAT2) in rat mammary gland Am J Physiol Endocrinol Metab, November 1, 2006; 291(5): E1059 - E1066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Ohnemus, M. Uenalan, J. Inzunza, J.-A. Gustafsson, and R. Paus The Hair Follicle as an Estrogen Target and Source Endocr. Rev., October 1, 2006; 27(6): 677 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Vallejo, C. Ballare, J. Lino Baranao, M. Beato, and P. Saragueta Progestin Activation of Nongenomic Pathways via Cross Talk of Progesterone Receptor with Estrogen Receptor {beta} Induces Proliferation of Endometrial Stromal Cells Mol. Endocrinol., December 1, 2005; 19(12): 3023 - 3037. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E E Connor, D L Wood, T S Sonstegard, A F da Mota, G L Bennett, J L Williams, and A V Capuco Chromosomal mapping and quantitative analysis of estrogen-related receptor alpha-1, estrogen receptors alpha and beta and progesterone receptor in the bovine mammary gland J. Endocrinol., June 1, 2005; 185(3): 593 - 603. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Ohnemus, M. Uenalan, F. Conrad, B. Handjiski, L. Mecklenburg, |
