| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
, but
not -
, in en face arterial endothelium
Department of Physiology, University of British Columbia, Vancouver, British Columbia, Canada V6T 1Z3
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Rapid, nongenomic effects of
17
-estradiol (E2) in endothelial cells are postulated to
arise from membrane-associated estrogen receptors (ERs), which have not
been visualized in vascular tissue. To identify membrane ERs, we used
multiple site-directed ER
or ER antibodies to label en face rat
cerebral and coronary arterial endothelia. Western blots revealed a
novel 55-kDa ER isoform. Three-dimensional images of cells labeled
with these antibodies and markers for the nucleus and caveolin-1 were
acquired with a wide-field microscope, deconvolved, and numerically
analyzed. We found ER in the nucleus and cell periphery, where
one-third colocalized with caveolin-1. The receptor location was
dependent on the epitope of the antibody. Human ovarian surface
epithelium produced similar results; but in rat myometrium, the
distribution was epitope independent and nuclear. ER distribution
was predominately intranuclear and epitope independent. A small amount
of ER colocalized with ER within the nucleus. The results were
identical in both arterial preparations and insensitive to
E2. We postulate that the different ER conformations at
the membrane, in the nucleus, and between different cell types allow
E2 to trigger cell- and location-specific signaling cascades.
estrogen receptors; immunocytochemistry; immunofluorescence; deconvolution; caveolin-1; colocalization
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
THE OVARIAN STEROID
estrogen [17-estradiol (E2)] induces both rapid
vasodilatory and long-term genomic effects in the vasculature, in part
by acting directly on the vascular endothelium (25). The
genomic effects are known to be mediated by specific estrogen receptors
(ERs) ER (13, 19) and ER (20, 28),
which function as ligand-activated transcription factors to modulate
the expression of vasoactive substances and adhesion molecules
(25). Studies using ER knockout mice demonstrate that both
receptor subtypes play important roles in the vasculature
(29).
The mechanism by which E2 evokes rapid,
endothelium-dependent vasodilation is not well understood. Studies
have shown that this effect does not involve transcription but does
require the ER. In cultured vascular endothelial cells, E2
acutely stimulates an ER-mediated production of nitric oxide (NO), a
potent vasodilator (32), through rapidly activating
endothelial NO synthase (eNOS) (5, 14, 39). Because this
response can be mimicked by membrane-impeded ligands (14,
37), such as BSA-conjugated E2, it has been proposed that there is a membrane ER. This argument is supported by endothelial cell fractionation data, which indicates that both ER and ER are
present in the plasmalemmal fractions containing caveolin-1 and eNOS
(4, 5). Despite this biochemical evidence, the bulk of
immunocytochemical data indicate that irrespective of whether it binds
a ligand, the classical ER is a nuclear protein (18, 44).
Receptors that are on or near the membrane and close to caveolin-1 have
not been visualized in vascular tissues or in native endothelial cells.
While the lack of supporting immunocytochemical data in native
endothelium is puzzling, it could be explained if either the putative
membrane receptor is not the classical ER (30) or if different epitopes are exposed in different cellular compartments. The
latter situation occurs with actin, which exposes different epitopes in
the nucleus and cytosol (12), and caveolin-1, which does
so in the cytosol and trans-Golgi network (9). Because cell fractionation has indicated that the classical ER might locate at
the plasmalemma, we hypothesized that the membrane ER was in a
different configuration and/or bound to different partners from its
nuclear counterpart. We examined ER distribution in en face rat
cerebral and coronary arterial endothelia using site-directed antibodies specific for ER and ER. Three-dimensional (3-D) images of cells labeled with these antibodies and markers for the nucleus and
caveolin-1 were acquired with a wide-field microscope, deconvolved, and
numerically analyzed. We found an abundance of extranuclear ER, a
significant fraction of which colocalizes with caveolin-1. Detecting
these receptors depended on the epitope to which the antibody was
directed. We also found that the E2-induced intranuclear rearrangement of ERs (16, 41) and translocation of
extranuclear ER into the nucleus (27) observed in
cultured cells did not occur in native endothelial cells.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
All chemicals and media were obtained from Sigma (St. Louis, MO) unless otherwise specified. Animal handling followed the guidelines of the Canadian Council of Animal Care.
Animal Treatment and Tissue Isolation
En face arteries. Two-week postovariectomized Sprague-Dawley rats (12 wk old, Charles River Laboratories; St. Constant, Quebec, Canada) were randomly assigned to three treatment groups: vehicle (0.1 ml peanut oil), E2 (physiological concentration) in vehicle (1 µg E2/100 g body wt), and E2 (pharmacological concentration) in vehicle (10 µg E2/100 g body wt). E2 was administered through a subcutaneous injection at the neck. Animals were anesthetized by intraperitoneal injection of pentobarbital sodium (60 mg/kg body wt) shortly before blood was extracted from the inferior vena cava for radioimmunoassay measurements of E2 (Coat-A-Count, DPC; Los Angeles, CA). Blood and tissue samples were taken at 15, 25, or 60 min after treatment with E2. The septal and basilar arteries were cut longitudinally to expose the endothelial layer and pinned to dissection trays as en face preparations in ice-cold phenol red-free HBSS (GIBCO-BRL; Burlington, Ontario, Canada) for further processing.
Uterine cryosections.
Uterine tissues were dissected from anesthetized cycling Sprague-Dawley
rats (12 wk old, Charles River Laboratories) during the proestrous
phase, as determined by the cytology of the animals' vaginal smears.
Dissected tissues were embedded in liquid Tissue-Tek (Sakura; Torrance,
CA) and flash frozen in 
60°C isopentane chilled in liquid
N2. Frozen tissues were cut transversely into 15-µm sections at 20°C using a cryostat, mounted onto slides, and stored at 80°C until being further processed for immunocytochemistry.
Cell Cultures
The IOSE-29 cell line was generated by transfecting normal human ovarian surface epithelial (OSE) cells in passage 5 with immortalizing simian virus 40 virus early genes (1). Cells were plated on glass coverslips in a 1:1 mixture of medium 199-MCDB 105 medium supplemented with 5% FBS (GIBCO-BRL). Two days before immunocytochemistry, cells were switched to phenol red-free media supplemented with serum replacement.Endothelial Uptake of Acetylated LDL
En face arteries were incubated with 10 µg/ml acetylated LDL labeled with 1,1'-dioctadecyl-3,3,3',3'-tetramethylindo-carbocyanine perchlorate (diI-Ac-LDL, BT-902, Biomedical Technologies; Stoughton, MA) in DMEM (GIBCO-BRL) at 37°C for 4 h, washed in PBS, and mounted onto glass coverslips for visualization using a narrow bandpass rhodamine filter (XF33, Omega Optical; Brattleboro, VT).Western Blot Analysis
En face arterial preparations were incubated in ice-cold lysis buffer (50 mM Tris, 150 mM NaCl, 20 mM EDTA, 0.5 mM PMSF, 2 µg/ml leupeptin, and 2 µg/ml aprotinin; pH 8.5, 4°C) for 45 min. Control tissues and the uterus and ovaries were homogenized in ice-cold lysis buffer. Protein concentration was measured with the BCA protein assay kit (Pierce; Rockford, IL). Samples to be labeled with SRA1000 or SRA1010 antibodies (names of antibodies, their sources, and applied concentrations are listed in Table 1) were further treated with benzonase nuclease (24 U/ml, Novagen; Madison, WI) at 37°C for 30 min. Tissue lysate, tissue homogenate, and recombinant ER (RP310, Affinity Bioreagents; Golden, CO) or
ER (RP311, Affinity Bioreagents) proteins were boiled with equal
amounts of sample buffer (0.125 M Tris · HCl, 4%
SDS, 20% glycerol, 10% mercaptoethanol, and 0.01% pyronin Y dye;
pH 6.8) for 5 min. Samples were loaded into a 5% stacking gel, 12.5%
running gel, and resolved using SDS-PAGE (21). Proteins
were transferred to nitrocellulose membranes, and the membranes were
then blocked with 5% nonfat dry milk in Tris-buffered saline-Tween
(TBS-T; 50 mM Tris, 0.09% NaCl, and 0.1% Tween; pH 7.6). The membrane was probed with an ER or ER antibody diluted in TBS-T with 1% BSA, rinsed with TBS-T, and incubated with the appropriate horseradish peroxidase (HRP)-conjugated antibody, either HRP-conjugated goat anti-mouse IgG or HRP-conjugated goat anti-rabbit IgG (both 1:10,000, Jackson ImmunoResearch Laboratory; West Grove, PA). Blots were rinsed
with TBS-T, and the signals were detected with the chemiluminescence Western blotting detection system (Amersham; Baie d'Urfe, Quebec, Canada).
|
Two sets of control experiments were performed to test the specificity of the reacted protein bands in blots. First, blots were incubated with ER antibodies that had been preabsorbed with their immunogenic peptides. Second, blots were incubated in antibody buffer without the primary antibody.
Fluorescent Immunocytochemistry
En face arteries, uterine sections, and human OSE cells were fixed for 10 min in freshly made 2% paraformaldehyde in PBS (in mM: 137 NaCl, 8 NaH2PO4, 2.7 KCl, and 1.5 KH2PO4; pH 7.4), quenched with 100 mM glycine (pH 7.4), permeabilized for 10 min in 0.1% Triton X-100, and washed in PBS. Fixed tissues and cells were blocked with antibody buffer (in mM: 75 NaCl, 18 Na3citrate, 2% goat serum, 1% BSA, 0.05% Triton X-100, and 0.02% NaN3) and incubated with one of the primary ER antibodies (Table 1). Arterial tissues were either labeled with one of the primary ER antibodies alone or in combination with a polyclonal caveolin-1 antibody (C13630, 1 µg/ml, Transduction Laboratories; Lexington, KY). Antibodies were diluted in antibody buffer and incubated with the tissues for 48 h at 4°C, washed with antibody wash solution (in mM: 75 NaCl abd 18 Na3 citrate with 0.05% Triton X-100), and rinsed in PBS. Tissues were incubated with the appropriate fluorophore-conjugated antibodies (4 µg/ml Alexa 488 goat anti-mouse or anti-rabbit IgG or 4 µg/ml Alexa 594 goat anti-mouse or anti-rabbit IgG, Molecular Probes; Eugene, OR), washed with antibody wash solution, and rinsed in PBS. To label the nuclei, tissues were incubated in 4',6-diamidino-2-phenylindole (DAPI; 300 nM, Molecular Probes) for 5 min and mounted onto coverslips in a solution composed of 90% glycerol, 10% 10× PBS, 2.5% triethylenediamene, and 0.02% NaN3. Tetraspeck fluorescent microspheres (0.56 µm, Molecular Probes) were added to the mounting medium as fiduciary markers, permitting accurate alignment of the 3-D images.Two sets of immunocytochemical control experiments were performed to test the specificity of the immunofluorescent label. First, tissues were incubated with ER antibodies that had been either preabsorbed with their respective immunogenic peptides or recombinant proteins. Second, tissues were incubated in antibody buffer without the primary antibody. Images of these experiments displayed only dim, diffuse, nonspecific staining.
Image Deconvolution and Analysis
Images were acquired with a Nikon Diaphot 200 inverted microscope equipped for epifluorescence (100-W Hg illumination). Low-magnification images were obtained with a 60/1.4 Planapo objective and a ×2 projection lens, giving pixel dimensions of 200 × 200 nm. High-magnification images were obtained with the same objective but with a ×4 projection lens, giving pixel dimensions of 100 × 100 nm. The former configuration allows a wide field of cells to be acquired simultaneously, whereas the latter configuration satisfies Nyquist criteria for sampling and prevents aliasing. The image detector was a thermoelectrically cooled charge-coupled device camera with a SiTe S1502AB chip and peak quantum efficiency of 80% (Photometrics). A series of two-dimensional images were acquired through the cells at 0.25-µm intervals. The point-spread function of the microscope was measured similarly by using fluospheres (100 nm diameter) of the appropriate color (Molecular Probes). Images were prepared as previously described (38) and then submitted to an EPR client-server processor for deconvolution (Scanalytics; Billerica, MA) using the algorithm developed by Carrington et al. (3).After deconvolution, the images were aligned using the fiduciary markers, and control images (primary antibody omitted from incubation) were used to identify the threshold intensity that eliminated >99% of the voxels in these images. This threshold intensity was applied to images of the fully labeled cells such that voxels that fell below the threshold were set to zero; all other voxels remained unchanged.
To determine whether an ER was intra- or extranuclear, the coordinates of the nuclear surface were determined by drawing a series of boundary lines to create a mesh around, but not touching, the nucleus. The mesh was then allowed to shrink until it contacted the surface (23). This resulted in a solid image, or mask, having the same shape as the nucleus, in which all voxels on or inside the mesh were given a value of one, whereas any voxel outside the surface was assigned a value of zero. A new image was created in which each voxel was a product of corresponding voxels in the ER image and the mask. This, in effect, was a copy of the ER image with all extranuclear signal set to zero, whereas the voxels located within the nuclear boundaries were unchanged. From the values in this image, we could then calculate how many above-threshold voxels were within the nucleus, and from the total number of illuminated voxels in the original image, we could calculate the number in the extranuclear compartment.
Statistical Analysis
The percentages of voxels located in the nucleus and colocalization measurements were expressed as means ± SD. Statistical significance was determined by the Mann-Whitney U-test (P < 0.05) using GraphPad Prism (GraphPad Software; San Diego, CA).| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Endothelial Integrity
En face arteries were incubated with the fluorescent derivative diI-Ac-LDL, which is an established marker of endothelial integrity (31). Fluorescence was present in the cytoplasm, but excluded from the nucleus, in all of the endothelial cells visualized, demonstrating that our techniques preserved the integrity and viability of the endothelial layer.ER Expression and Antibody Specificity
ER expression and antibody specificity were determined by Western blots of septal and basilar artery lysates, uterine and ovarian homogenates, and recombinant human ER and ER proteins (Fig.
1). The immunogenic sequences, sources,
and concentrations of the antibodies are listed in Table 1.
|
The NH2-terminus (Fig. 1A), hinge region (Fig.
1B), and COOH-terminus (Fig. 1C) ER antibodies
all detected the same two bands of comparable optical density in septal
(lane 1) and basilar (lane 2) artery lysates and
uterine homogenate (lane 3). The heavier band migrated at
~67 kDa, and the lighter band migrated at ~55 kDa. Receptor subtype
specificity was determined using recombinant ER (lane 4)
and ER (lane 5) proteins. All the antibodies reacted strongly with ER at 67 kDa but failed to recognize a 10-fold higher
concentration of ER.
The three ER antibodies, NH2 terminus (Fig.
1D), ligand-binding domain (LBD; Fig. 1E), and
COOH terminus (Fig. 1F), detected the same two bands at
comparable intensities in septal (lane 1) and basilar artery
(lane 2) lysates and in ovarian homogenate (lane
3). The heavier band migrated at ~63 kDa, and the lighter band
migrated at ~54 kDa. The ER antibodies recognized the long-form recombinant ER protein at the expected molecular mass of 63 kDa (lane 5) but did not detect an almost sevenfold higher
concentration of ER (lane 4).
Blots labeled with antibodies preadsorbed with their immunizing peptides or with secondary antibodies alone did not show any bands.
Serum E2
Radioimmunassay measurements showed that the serum E2 concentration of vehicle-treated ovariectomized rats was 18 ± 3 pg/ml (n = 6). In vivo treatment with a physiological concentration of E2, 1 µg/100 g body wt, increased the serum level of E2 to 383 ± 154 pg/ml (n = 6), which is comparable to that observed in cycling rats (26). Treatment with a pharmacological concentration of E2, 10 µg/100 g body wt, raised the serum level of E2 to 7,721 ± 1,823 pg/ml.ER in Coronary Artery Endothelia
NH2 terminus (ER21).
Figure 2, A and C,
display low-magnification, raw and deconvolved images, respectively, of
the ER distribution detected with ER21 in the septal artery
endothelium of a vehicle-treated animal. Corresponding images of the
nuclear stain DAPI are shown in Fig. 2, B and D.
Images that have been deconvolved show higher contrast and resolution
and provide considerably more information about receptor distribution
than do conventional images. The large surface area that can be viewed
at low magnification demonstrates that this antibody labeled all of the
endothelial cells and suggests that the receptors are distributed
primarily within the nucleus. These images are displayed with the long
axis of the vessels running almost parallel to the y-axis.
The nuclei of the cells are oriented with their long axis in the
direction of the blood flow, which is characteristic of native vascular
endothelium (2).
|
and the nuclear mask allowed us to accurately determine
the percentage of the ER within the nucleus, 85 ± 3% in the
four cells examined (Table 2).
Experiments (not shown) demonstrate that there is no difference between
the boundary of the nucleus measured with this technique and the
boundary visualized using an antibody against nuclear pore complex
proteins (Mab414, BABCO; Richmond, CA).
|
within the nucleus, the
solid mask was hollowed out to produce a shell marking the nuclear
boundary and then merged with the image of the ER. The resulting
image was either bisected in x-y plane (Fig.
2I) or rotated about the y-axis and bisected in
the y-z plane (Fig. 2J). These images
show that the nuclear receptors are distributed as distinct clusters
scattered throughout the volume of the nucleus with the few
extranuclear receptors being mainly in the perinuclear space.
When ER21 was preneutralized, the staining is dim, diffuse, and
nonspecific (Fig. 2K). Comparable results were obtained with all of the preneutralized ER antibodies and when cells were labeled without primary antibody.
COOH terminus (SRA1010).
Low-magnification images (Fig. 3, A and
B) show that, like ER21, this
antibody labeled all of the septal artery endothelial cells examined in
a vehicle-treated animal, but the distribution pattern was markedly
different. The high-magnification image (Fig. 3C,i) shows that ER appears as small, punctate
clusters whose pattern of distribution mirrors the narrow and elongated
cell shape characteristic of native vascular endothelium
(11). Superimposing this image with its nuclear shell and
bisecting it in the x-y plane (Fig.
3C,ii) revealed that the voxels containing
signals specific for the COOH terminus of the ER were largely
excluded from the nucleus. Specific label could be seen at the cell
periphery, in the perinuclear region, and within the cytoplasm. The
numerical analysis indicated that only 4 ± 2% of the labeled
voxels were within the nuclear boundary (Table 2). A side view (Fig.
3C,iii) of the cell shows extranuclear ER
located in both the apical and basolateral regions, above and below the
nucleus, where they could be on or near the cell membrane.
|
Hinge region (SRA1000). Similar images were acquired when the septal artery of vehicle-treated animals was labeled with an antibody directed against the hinge. Every endothelial cell was labeled, and the high-magnification images indicated a predominantly extranuclear distribution (Fig. 3D,i). When combined with the nuclear mask, digital analysis showed that the intranuclear label accounted for only 8% of the labeled voxels in this image (Fig. 3D,ii), and just 7 ± 2% of all of the voxels labeled with this antibody (n = 4, Table 2). A side view shows that the extranuclear label was distributed on both the apical and basolateral surfaces (Fig. 3D,iii).
ER in Cells of Reproductive Organs
was present in other cell types as well as in other species. We
chose to examine human OSE cells and rat myometrial smooth muscle cells
because they are important targets of E2 and are known to
express ER (6, 42). Figure
4 displays high-magnification, merged,
bisected (in the x-y plane) images of ER21
(A) or SRA1010 (B) and the corresponding nuclear
mask in human OSE cells. These data show that, similar to the coronary
artery endothelia, ER21 detected mainly nuclear receptors, whereas
SRA1010 detected predominately extranuclear receptors. The SRA1000
antibody also produced predominately extranuclear label. In contrast,
high-magnification images of rat myometrial longitudinal smooth muscle
cells dual labeled with ER21 (Fig.
5A) and DAPI (Fig.
5B), or SRA1010 (Fig. 5C) and DAPI (Fig.
5D), show that the ER distribution is predominately
nuclear irrespective of the antibody epitope. The SRA1000 antibody did not label these cells.
|
|
ER in Coronary Artery Endothelia
in the endothelium of septal arteries.
Figure 6, A and B,
shows low-magnification images of the ER distribution labeled with
an antibody directed against the LBD and the corresponding DAPI
staining in the endothelium of a vehicle-treated animal. All cells
lining the luminal surface showed the presence of ER, which at high
magnification (Fig. 6C) appeared to be predominately nuclear. Figure 6D displays the image merged with the
corresponding image of the nuclear shell and confirms that the vast
majority of voxels are located within the nucleus. This image and a
side view of the cell bisected in the y-z plane
(Fig. 6E) show that the receptors are distributed, like
those detected with ER21, as distinct clusters dispersed throughout the
nucleus. On average, 88 ± 3% of voxels labeled with LBD antibody
were intranuclear (Table 2). Similar images of the ER distribution
were recorded with an antibody against the COOH terminus (Table 2). We
tested three commercially available antibodies targeted against various segments of the A/B region but found no labeling (PA1311, Affinity Bioreagents; Y19, Santa Cruz Biotechnologies; 06629, Upstate), although
they all reacted with two bands of 63 and 54 kDa in Western blots.
|
Extranuclear ER in Coronary Artery Endothelia
that we detected with antibodies against the
COOH terminus or hinge region of the molecule were extranuclear, and
many appeared to be close to the surface (Fig. 3,
C,iii and D,iii). Because
immediate responses of E2 have been linked to the
ER-mediated activation of eNOS within caveolae, we tested whether
extranuclear ER was positioned in close proximity to caveolin-1, a
marker for caveolae in endothelial cells (33).
In Fig. 7A, we show a
stereopair of the distribution of caveolin-1 (red) and the COOH
terminus of ER (green) in a septal artery endothelial cell of a
vehicle-treated animal. Caveolin-1 was distributed in distinct clusters
at the periphery of the cell, but the clusters do not form a continuous
surface, in agreement with earlier results (33). The large
number of white voxels demonstrates that there is considerable
colocalization between the two proteins, although not every caveolin-1
is associated with an ER. In this and other cells examined, 32 ± 3% of the voxels that were labeled with the COOH-terminus antibody
also contained caveolin-1 (Table 3).
Comparable levels of colocalization were recorded with the antibody
directed against the hinge region (Table 3). These values are
significantly greater than zero (P < 0.005).
|
|
Intranuclear ER and ER in Coronary Artery Endothelia
and ER, we tested the
hypothesis that the two receptors are in positions where they could interact. We visualized the distribution of ER using the
NH2-terminus (ER21) antibody and ER using the LBD
(CO1531) antibody in the same cells. The results (Fig. 7B)
indicate that a small (14 ± 2%, n = 4) number of
voxels containing ER are colocalized with ER. This number is
significantly greater than zero (P < 0.005).
Basilar Artery
Next, we investigated whether the epitope-dependent distribution of ER and the epitope-independent nuclear distribution of ER were
present in the endothelia of other vascular beds. We chose to examine
the basilar artery because E2 is known to have important
effects in the cerebral vasculature (34). We found that
the distribution patterns of ER and ER in the basilar artery endothelium are virtually identical to that observed in the coronary artery (Table 2), as was the colocalization of ER with caveolin-1 (Table 3) and the colocalization of ER with ER (12 ± 3%,
n = 4).
Effect of E2
Studies have shown that unliganded, cytoplasmic, steroid receptors (44), including ERs (27), can undergo nuclear translocation and that nuclear ERs redistribute within the nucleus (16, 41) after E2 exposure. Because we observed an abundance of both extranuclear and nuclear ER, we
investigated whether these phenomenon occur in the vascular endothelium
using comparable time points as that used in previous studies
(16, 27, 41).
In vivo treatment with physiological concentrations of E2
for 30 min did not alter the pattern of ER distribution recorded with
the five antibodies (Table 2). Additional experiments measuring the
effects of E2 at 15 and 60 min and with pharmacological
concentrations (data not shown) also had no apparent effect. The values
obtained from the physiological and pharmacological treatments were
virtually identical. The colocalization of ER with ER and the
colocalization of ER with caveolin-1 (Table 3) were similarly
unaffected. Identical results were obtained in both septal and basilar arteries.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Antibody Specificity
We have verified that the ER antibodies used in this study were either ER or ER specific using a number of approaches. First, BLAST searches in the NCBI protein data bank showed that the sequences used as immunogens (Table 1) matched ER and no other proteins. Second,
all Western blots using ER antibodies were identical, as were the
blots using ER antibodies, producing bands of the expected molecular
weight regardless of the epitope the antibody recognized. ER
antibodies reacted with recombinant ER protein but not recombinant
ER protein, and vice versa (Fig. 1). Antibodies preneutralized with
their immunogenic peptide, or with recombinant protein, produced no
bands in the Western blots. Blots probed only with secondary antibody
were also blank. Finally, immunocytochemical experiments using either
preneutralized antibodies, or only the secondary antibody, produced
dim, diffuse labeling (Fig. 2K). Taken together, these
results indicate that the primary antibodies are specific for a given
isoform of the ER and that the secondary antibodies are binding only to
the primary antibodies.
ER Isoforms
Our Western blots (Fig. 1, A-C) show that ER is
expressed in rat vascular and uterine tissues as two isoforms at 67 and
55 kDa. The heavier band has the expected molecular weight of the rat
ER (19). The lighter band is not a breakdown product of the 67 kDa ER, because the epitopes recognized by the ER
antibodies, in the NH2 terminus, COOH terminus, and hinge
regions, were all detected. The Western blots eliminate the possibility
that this is a shorter transcript arising from translation initiated at the sixth or seventh ATG because the first 21 amino acids (the epitope
for ER21) would be missing (15). The intact epitope in the
hinge region eliminates the possibility that this is ER
4 (17,
40), which is missing the entire hinge region, leaving a
previously unreported isoform as the most likely explanation for the
lighter band.
We found that the rat vascular and ovarian tissues expressed two
isoforms of ER, at 54 and 63 kDa (Fig. 1,
D-F). The lighter band represents either the
485-amino acid ER originally cloned in the rat ovary
(20) or the 503-amino acid rat ER with an 18-amino acid
in-frame insertion in the LBD (43). The heavier band has
the same molecular weight as the long-form rat ER, which has 549 amino acids due to a 64-amino acid NH2-terminus extension (22). These previously reported ER isoforms differ in
either their ligand binding affinity or rate of transcriptional activity.
Epitope-Dependent Localization of ER
expose different epitopes from their intranuclear
(Fig. 2) counterparts. The masking of one epitope and unmasking of
another could result from a conformational change in the protein
(12), from the protein being associated with different
partners (10), or a combination of both. This view is
further supported by our Western blots. After the ER proteins were
linearized and their binding partners removed (Fig. 1), all of the
epitopes became accessible, and all three ER antibodies detected
bands with comparable molecular weights. In addition, we found that
human OSE cells exhibit virtually identical receptor distribution
patterns, demonstrating that the epitope-dependent localization was
neither restricted to vascular endothelial cells nor to the rat
species. It is, however, cell specific, because both ER21 and SRA1010
antibodies detected predominately nuclear receptors in rat myometrial
smooth muscle cells, whereas SRA1000 produced no label. These data show
that ER takes on different conformations and/or binding partners
between nuclear and extranuclear compartments and between cell types.
While it is well known that the binding of different ligands alters
ER conformation differentially (24), we have observed a
number of different conformations in the ligand-free state. Having
distinct ER conformations may be one of the ways in which cells can
respond differently to the same hormone via the same receptor protein.
Extranuclear ER and Caveolae
colocalizes with caveolin-1, an essential component of endothelial caveolae (33), confirming that these ER are located in
or near the plasmalemma. Recent cell fractionation studies showed that a subpopulation of ER is associated with the endothelial caveolae and that ER in the caveolar fraction can acutely activate eNOS when
treated with E2 (5). Our colocalization data
confirm that ER are associated with the caveolae, and this is the
first such demonstration in intact native vascular endothelium. These
data, combined with the Western blots, suggest that the plasma membrane ER is identical to the classical ER. Furthermore, our work shows that the extranuclear receptors have configurations and/or binding partners different from their nuclear counterparts. This provides a
model that explains how the same ER can mediate both the rapid, nontranscriptional, and long-term transcriptional effects of
E2.
The finding of an abundance of extranuclear ER in arterial
endothelia, as well as in human OSE cells, contradicts the established view that both liganded and unliganded ER are largely nuclear proteins (44). Some extranuclear protein is expected,
given the nucleocytoplasmic shuttling hypothesis (8), but
our observations can only be explained with a model in which ER is
both a nuclear and membrane- associated molecule.
ER Distribution
antibodies produced positive immunoreactivity in all the
arterial endothelial cells that we examined. Unlike ER, ER is
predominately nuclear, irrespective of the antibody we used. Antibodies
against the A/B region produced clear Western blots but no
immunocytochemical staining, suggesting that the A/B region is
inaccessible in vivo. Because there are so few extranuclear ER, it
is unlikely to have any involvement in mediating the rapid, nontranscriptional effects of E2. This view agrees with
recent work showing that ER, but not ER, can acutely activate
phosphotidylinositol-3-kinase and eNOS in cultured human vascular
endothelial cells (39). However, a recent study using
cellular fractionation has demonstrated that ER is present in
cultured ovine endothelial caveolae and can rapidly activate eNOS when
exposed to E2 (4). This discrepancy may be due
to culturing or species difference or because the receptors cannot be
detected by the antibodies used in our study.
Colocalization of ER and ER
and ER can form
heterodimers both in vitro and in transfected cells (7,
35), but the relative receptor distribution in nuclei with an
intact molecular architecture is unknown. We found that nuclear ER
and ER share a statistically significant degree of colocalization,
although most of the receptors are not located in close proximity.
These data indicate that a small proportion of the receptor subtypes are in positions where they can heterodimerize, and this is the first
such demonstration in an intact cell. In addition, the data suggest
that the majority of the two receptors have distinct transcriptional functions in the vascular endothelium.
Effects of E2
An unexpected result is the lack of a detectable effect of estradiol on the distribution of the receptors regardless of the antibody used to visualize the receptor. Because the serum E2 concentration of the vehicle-treated animals was 18 pg/ml (0.07 pmol/l), well below the dissociation constants of ER
(~0.3 nmol/l) (36) and ER (~0.6 nmol/l)
(20), it is likely that the receptors in these animals
were ligand free. We also tested the distribution of the receptors in
cycling animals, in ovariectomized animals treated with pharmacological
dosages of E2 (10 µg E2/100 g body wt,
radioimmunoassay = 7,721 ± 1,823 pg/ml), and in
ovariectomized animals treated with E2 for 15 and 60 min.
In all cases, E2 had no observable or measurable effect on
distribution or colocalization.
Our results are contrary to those observed in green fluorescent protein
(GFP)-ER-transfected cells (16, 41) in which the
protein was only located within the nucleus and formed clumps in
response to E2. The discrepancies may arise from a number
of sources. First, the cell type and experimental conditions are completely different from the two previous studies, which used HeLa or
human breast cancer cells with overexpressed GFP-conjugated receptors
versus endogenous ER in native arterial tissues examined in our
studies. Second, in the absence of the amplifying effect of
overexpression, the changes of intensity and location of the ligand-bound receptors may have been too small for us to detect.
This study shows that certain cell types have abundant extranuclear
ER, some of which colocalize with caveolae and may represent the
membrane-associated receptor. The discovery that these receptors have a
different configuration from their nuclear counterparts suggests that
they may be involved in pathways different from those of the nuclear
receptor and warrants further investigation.
| |
ACKNOWLEDGEMENTS |
|---|
The authors express gratitude to Dr. Geoffrey L. Greene (University of Chicago, Chicago, IL) for the kind donations of ER21 and CO1531 antibodies and to Dr. C. Roskelley (University of British Columbia, British Columbia, Canada) for the donation of IOSE-29 cells.
| |
FOOTNOTES |
|---|
This work was funded by a Doctoral Research Award from the Heart and Stroke Foundation of Canada (to P. Dan), a summer student fellowship from the Heart and Stroke Foundation of British Columbia and the Yukon (to J. C. Y.Cheung), and a Grant-in-Aid from the Heart and Stroke Foundation of British Columbia and the Yukon (to E. D. W. Moore).
Address for reprint requests and other correspondence: E. D. W. Moore, Dept. of Physiology, Univ. of British Columbia, 2146 Health Sciences Mall, Vancouver, British Columbia, Canada V6T 1Z3 (E-mail: edmoore{at}interchange.ubc.ca).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
First published December 12, 2002;10.1152/ajpheart.00781.2002
Received 5 September 2002; accepted in final form 4 December 2002.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Auersperg, N,
Pan J,
Grove BD,
Peterson T,
Fisher J,
Maines-Bandiera S,
Somasiri A,
and
Roskelley CD.
E-cadherin induces mesenchymal-to-epithelial transition in human ovarian surface epithelium.
Proc Natl Acad Sci USA
96:
6249-6254,
1999
2.
Ballermann, BJ,
Dardik A,
Eng E,
and
Liu A.
Shear stress and the endothelium.
Kidney Int Suppl
67:
S100-S108,
1998[Medline].
3.
Carrington, WA,
Lynch RM,
Moore EDW,
Isenberg G,
Fogarty KE,
and
Fay FS.
Superresolution three-dimensional images of fluorescence in cells with minimal light exposure.
Science
268:
1483-1487,
1995
4.
Chambliss, KL,
Yuhanna IS,
Anderson RG,
Mendelsohn ME,
and
Shaul PW.
ERbeta has nongenomic action in caveolae.
Mol Endocrinol
16:
938-946,
2002
5.
Chambliss, KL,
Yuhanna IS,
Mineo C,
Liu P,
German Z,
Sherman TS,
Mendelsohn ME,
Anderson RGW,
and
Shaul PW.
Estrogen receptor and endothelial nitric oxide synthase are organized into a functional signaling module in caveolae.
Circ Res
87:
E44-E52,
2000[Web of Science][Medline].
6.
Choi, KC,
Kang SK,
Tai CJ,
Auersperg N,
and
Leung PC.
Estradiol up-regulates antiapoptotic Bcl-2 messenger ribonucleic acid and protein in tumorigenic ovarian surface epithelium cells.
Endocrinology
142:
2351-2360,
2001
7.
Cowley, SM,
Hoare S,
Mosselman S,
and
Parker MG.
Estrogen receptors alpha and beta form heterodimers on DNA.
J Biol Chem
272:
19858-19862,
1997
8.
Dauvois, S,
White R,
and
Parker MG.
The antiestrogen ICI 182780 disrupts estrogen receptor nucleocytoplasmic shuttling.
J Cell Sci
106:
1377-1388,
1993[Abstract].
9.
Dupree, P,
Parton RG,
Raposo G,
Kurzchalia TV,
and
Simlns K.
Caveolae and sorting in the trans-Golgi-network of epithelial cells.
EMBO J
12:
1597-1605,
1993[Web of Science][Medline].
10.
Dyer, J, IRK,
Pugh G,
Quinlan R,
Lane E,
and
Hutchison C.
Cell cycle changes in A-type lamin associations detected in human dermal fibroblasts using monoclonal antibodies.
Chromosome Res
5:
383-394,
1997[Web of Science][Medline].
11.
Gabaldon, M,
and
Marques A.
Effect of inflammatory stimuli on the silver staining pattern of the rat carotid endothelium.
Histochem J
31:
153-160,
1999[Web of Science][Medline].
12.
Gonsior, S,
Platz S,
Buchmeier S,
Scheer U,
Jockusch B,
and
Hinssen H.
Conformational difference between nuclear and cytoplasmic actin detected by a monoclonal antibody.
J Cell Sci
112:
797-809,
1999[Abstract].
13.
Green, S,
Walter P,
Kumar V,
Krust A,
Bornert JM,
Argos P,
and
Chambon P.
Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A.
Nature
320:
134-139,
1986[Medline].
14.
Haynes, MP,
Sinha D,
Russell KS,
Collinge M,
Fulton D,
Morales-Ruiz M,
Sessa WC,
and
Bender JR.
Membrane estrogen receptor engagement activates endothelial nitric oxide synthase via the PI3-kinase-Akt pathway in human endothelial cells.
Circ Res
87:
677-682,
2000
15.
Hiroi, H,
Inoue S,
Watanabe T,
Goto W,
Orimo A,
Momoeda M,
Tsutsumi O,
Taketani Y,
and
Muramatsu M.
Differential immunolocalization of estrogen receptor alpha and beta in rat ovary and uterus.
J Mol Endocrinol
22:
37-44,
1999[Abstract].
16.
Htun, H,
Holth LT,
Walker D,
Davie JR,
and
Hager GL.
Direct visualization of the human estrogen receptor alpha reveals a role for ligand in the nuclear distribution of the receptor.
Mol Biol Cell
10:
471-486,
1999
17.
Karas, RH,
Baur WE,
van Eickles M,
and
Mendelsohn ME.
Human vascular smooth muscle cells express an estrogen receptor isoform.
FEBS Lett
377:
103-108,
1995[Web of Science][Medline].
18.
Kim-Schulze, S,
McGowan KA,
Hubchak SC,
Cid MC,
Martin MB,
Kleinman HK,
Greene GL,
and
Schnaper HW.
Expression of an estrogen receptor by human coronary artery and umbilical vein endothelial cells.
Circulation
94:
1402-1407,
1996
19.
Koike, S,
Sakai M,
and
Muramatsu M.
Molecular cloning and characterization of rat estrogen receptor cDNA.
Nucleic Acids Res
15:
2499-2513,
1987
20.
Kuiper, GG,
Enmark E,
Pelto-Huikko M,
Nilsson S,
and
Gustafsson JA.
Cloning of a novel receptor expressed in rat prostate and ovary.
Proc Natl Acad Sci USA
93:
5925-5930,
1996
21.
Laemmli, UK.
Cleavage of structural proteins during the assembly of the head of bacteriophage T4.
Nature
227:
680-685,
1970[Medline].
22.
Leygue, E,
Dotzlaw H,
Lu B,
Glor C,
Watson PH,
and
Murphy LC.
Estrogen receptor beta: mine is longer than yours?
J Clin Endocrinol Metab
83:
3754-3755,
1998
23.
Lifshitz, LM,
Collins JA,
Moore EDW,
and
Gauch J.
Computer Vision and Graphics in Fluorescence Microscopy. Los Alamitos, CA: IEEE Computer Society Press, 1994.
24.
McDonnell, DP.
The molecular pharmacology of SERMs.
Trends Endocrinol Metab
10:
301-311,
1999[Web of Science][Medline].
25.
Mendelsohn, ME,
and
Karas RH.
The protective effects of estrogen on the cardiovascular system.
N Engl J Med
340:
1801-1811,
1999
26.
Menjivar, M,
Cardenas M,
Ortiz G,
and
Pedraza-Chaverri J.
Fertility diminution in female rats with experimental chronic nephrosis.
Biol Reprod
63:
1549-1554,
2000
27.
Monje, P,
Zanello S,
Holick M,
and
Boland R.
Differential cellular localization of estrogen receptor alpha in uterine and mammary cells.
Mol Cell Endocrinol
181:
117-129,
2001[Web of Science][Medline].
28.
Mosselman, S,
Polman J,
and
Dijkema R.
ER beta: identification and characterization of a novel human estrogen receptor.
FEBS Lett
392:
49-53,
1996[Web of Science][Medline].
29.
Mueller, SO,
and
Korach KS.
Estrogen receptors and endocrine diseases: lessons from estrogen receptor knockout mice.
Curr Opin Pharmacol
1:
613-619,
2001[Medline].
30.
Nadal, A,
Diaz M,
and
Valverde MA.
The estrogen trinity: membrane, cytosolic, and nuclear effects.
News Physiol Sci
16:
251-255,
2001
31.
Netland, PA,
Zetter BR,
Via DP,
and
Voyta JC.
In situ labelling of vascular endothelium with fluorescent acetylated low density lipoprotein.
Histochem J
17:
1309-1320,
1985[Web of Science][Medline].
32.
Palmer, RM,
Ferrige AG,
and
Moncada S.
Nitric oxide release accounts for the biological activity of endothelium-derived relaxing factor.
Nature
327:
524-526,
1987[Medline].
33.
Parton, RG.
Caveolae and caveolins.
Curr Opin Cell Biol
8:
542-548,
1996[Web of Science][Medline].
34.
Pelligrino, DA,
and
Galea E.
Estrogen and cerebrovascular physiology and pathophysiology.
Jpn J Pharmacol
86:
137-158,
2001[Medline].
35.
Pettersson, K,
Delaunay F,
and
Gustafsson JA.
Estrogen receptor beta acts as a dominant regulator of estrogen signaling.
Oncogene
19:
4970-4978,
2000[Web of Science][Medline].
36.
Razandi, M,
Pedram A,
Greene GL,
and
Levin ER.
Cell membrane and nuclear estrogen receptors (ERs) originate from a single transcript: studies of ERalpha and ERbeta expressed in Chinese hamster ovary cells.
Mol Endocrinol
13:
307-319,
1999
37.
Russell, KS,
Haynes MP,
Sinha D,
Clerisme E,
and
Bender JR.
Human vascular endothelial cells contain membrane binding sites for estradiol, which mediate rapid intracellular signaling.
Proc Natl Acad Sci USA
97:
5930-5935,
2000
38.
Scriven, DR,
Dan P,
and
Moore EDW
Distribution of proteins implicated in excitation-contraction coupling in rat ventricular myocytes.
Biophys J
79:
2682-2691,
2000[Web of Science][Medline].
39.
Simoncini, T,
Hafezi-Moghadam A,
Brazil DP,
Ley K,
Chin WW,
and
Liao JK.
Interaction of oestrogen receptor with the regulatory subunit of phosphatidylinositol-3-OH kinase.
Nature
407:
538-541,
2000[Medline].
40.
Skipper, JK,
Young LJ,
Bergeron JM,
Tetzlaff MT,
Osborn CT,
and
Crews D.
Identification of an isoform of the estrogen receptor messenger RNA lacking exon four and present in the brain.
Proc Natl Acad Sci USA
90:
7172-7175,
1993
41.
Stenoien, DL,
Mancini MG,
Patel K,
Allegretto EA,
Smith CL,
and
Mancini MA.
Subnuclear trafficking of estrogen receptor-alpha and steroid receptor coactivator-1.
Mol Endocrinol
14:
518-534,
2000
42.
Wang, H,
Masironi B,
Eriksson H,
and
Sahlin L.
A comparative study of estrogen receptors alpha and beta in the rat uterus.
Biol Reprod
61:
955-964,
1999
43.
Warner, M,
Nilsson S,
and
Gustafsson JA.
The estrogen receptor family.
Curr Opin Obstet Gynecol
11:
249-254,
1999[Web of Science][Medline].
44.
Yamashita, S.
Localization and functions of steroid hormone receptors.
Histol Histopathol
13:
255-270,
1998[Web of Science][Medline].
This article has been cited by other articles:
|
|
 |
|
  P. Galluzzo, C. Rastelli, P. Bulzomi, F. Acconcia, V. Pallottini, and M. Marino 17{beta}-Estradiol regulates the first steps of skeletal muscle cell differentiation via ER-{alpha}-mediated signals Am J Physiol Cell Physiol, November 1, 2009; 297(5): C1249 - C1262. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. Patel and R. R. Arora Review: Endothelial dysfunction: A potential tool in gender related cardiovascular disease Therapeutic Advances in Cardiovascular Disease, April 1, 2008; 2(2): 89 - 100. [Abstract] [PDF]
|
|
|
|
 |
|
 D. N. Krause, S. P. Duckles, and D. A. Pelligrino Influence of sex steroid hormones on cerebrovascular function J Appl Physiol, October 1, 2006; 101(4): 1252 - 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Guzzo, M. Salih, E. D. Moore, and B. S. Tuana Molecular properties of cardiac tail-anchored membrane protein SLMAP are consistent with structural role in arrangement of excitation-contraction coupling apparatus Am J Physiol Heart Circ Physiol, April 1, 2005; 288(4): H1810 - H1819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. X. Liao, R. R. Magness, and D.-b. Chen Expression of Estrogen Receptors-{alpha} and -{beta} in the Pregnant Ovine Uterine Artery Endothelial Cells In Vivo and In Vitro Biol Reprod, March 1, 2005; 72(3): 530 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Orshal and R. A. Khalil Gender, sex hormones, and vascular tone Am J Physiol Regulatory Integrative Comp Physiol, February 1, 2004; 286(2): R233 - R249. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. S. Watson and B. Gametchu Proteins of Multiple Classes May Participate in Nongenomic Steroid Actions Experimental Biology and Medicine, December 1, 2003; 228(11): 1272 - 1281. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
