| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Article |
and -ß
Centre for Women’s Health Research (C.E.G., M.Z., K.B., P.A.W.R.), Monash University Department of Obstetrics and Gynaecology, Monash Medical Centre, Clayton, Victoria 3168, Australia; and Prince Henry’s Institute of Medical Research (S.C., P.J.F.), Clayton, Victoria 3168, Australia
Address all correspondence and requests for reprints to: Dr. Caroline Gargett, Centre for Women’s Health Research, Monash University Department of Obstetrics and Gynaecology, Monash Medical Centre, 246 Clayton Road, Clayton, Victoria 3168, Australia. E-mail: . caroline.gargett{at}med.monash.edu.au
Abstract
Estrogen has a cardiovascular protective role in women due in part to its effect on the vasculature. The roles of the two estrogen receptors (ERs), ER
and ERß, in the vascular actions of estrogen are unclear, as are effects of estrogen on microvascular endothelial cells (MEC) derived from sex steroid-responsive tissues. The present study demonstrates that 17ß-estradiol, but not progesterone, increases vascular endothelial growth factor (VEGF) receptor (VEGFR) expression on human myometrial MEC measured using biotin-recombinant human (rh) VEGF165 and flow cytometry. This response occurred in a time- and dose-dependent manner, with significantly increased rhVEGF165 binding at 3 h and maximal responses between 0.1 and 10 nmol/liter 17ß-estradiol, which was blocked by the antiestrogen ICI 182,780. Approximately 60% of samples demonstrated this response to 17ß-estradiol. All samples of myometrial MEC expressed both ERß mRNA and protein demonstrated by semiquantitative RT-PCR and Western blotting. However, ER mRNA and protein were expressed in only 13 of 21 MEC samples. There was a significant association between ER expression in myometrial MEC and their ability to respond to 17ß-estradiol by increasing rhVEGF165 binding. 17ß-estradiol increased VEGFR-2 expression in ER-expressing MEC isolates, which also demonstrated increased rhVEGF165 binding, but failed to have these effects on ER negative samples. Similarly, 17ß-estradiol augmented VEGF-induced MEC proliferation in ER-expressing MEC samples, which was blocked by ICI 182,780. These observations suggest that 17ß-estradiol increases VEGFR-2 expression on human myometrial MEC promoting endothelial cell proliferation, an effect that varies between subjects and appears to be mediated primarily by ER.
THE UTERUS IS a classical sex steroid target organ, where both myometrial and endometrial cells express estrogen and progesterone receptors (ERs and PRs). The vascular system is also estrogen responsive (1, 2). Experimental studies show that 17ß-estradiol exerts a range of direct effects on endothelial cells (EC), which can be either rapid or long-term (3). These include up-regulation of endothelial nitric oxide synthase activity, modulation of adhesion molecule expression, and promotion of EC survival and angiogenic activity (3, 4, 5, 6, 7). In contrast, progesterone has proapoptotic effects and inhibits EC proliferation (8). Because ER and PR have been demonstrated in human large vessel EC (7, 8, 9) and in endometrial MEC (10), it is likely that sex steroid effects are mediated by these receptors, although estrogen also has nongenomic effects (11).
The two ERs, ER and ERß, which are encoded by separate genes, are involved in the vascular actions of 17ß-estradiol, but their relative roles are unclear (1, 12). ER and ERß bind 17ß-estradiol with similar high affinity causing receptor dimerization. They act as ligand-activated transcription factors for a range of estrogen target genes, binding to estrogen response elements or indirectly by interaction with other DNA binding proteins (13). However, ER and ERß can mediate opposing transcriptional activities, dependent on the type of response element in target gene promoters and other cell specific factors such as presence or absence of coregulators (12). There are also differences in the tissue distribution of ER and ERß, although there is considerable overlap and heterodimerization may occur in cells where they are coexpressed.
Vascular endothelial growth factor (VEGF) is a specific mitogen for EC, acting through two tyrosine-kinase receptors, VEGF receptor (VEGFR)-1 (flt-1) and VEGFR-2 (KDR), expressed almost exclusively on EC (14). Both receptors bind VEGF with high affinity, but signaling through VEGFR-2, rather than VEGFR-1, mediates the angiogenic effects of VEGF (14). Both VEGFRs are expressed at low levels in vivo and in quiescent cultured EC (15, 16) where they serve to promote EC survival (17) but are up-regulated during periods of angiogenesis (18). The factors regulating VEGFR expression are not fully understood, although hypoxia and some cytokines and growth factors may be involved (14).
17ß-estradiol is known to modulate the expression of some growth factors and their receptors (19). Because 17ß-estradiol up-regulates basic fibroblast growth factor (bFGF)-induced angiogenesis (6, 7, 10) we postulated that 17ß-estradiol and progesterone would also modulate VEGF-induced angiogenesis through alteration of VEGFR expression. While the study of sex steroid effects on endothelium has largely focused on large vessels, there are few studies examining the effects of 17ß-estradiol and progesterone on MEC derived from classic sex steroid-responsive tissues such as the uterus. We have used MEC isolated from the human myometrium as an in vitro model to investigate the effects of estrogen on the microvascular system. Here we demonstrate that 17ß-estradiol up-regulates VEGFR-2 expression in myometrial MEC. We also show that myometrial MEC express both ER and ERß and that 17ß-estradiol up-regulation of VEGFR-2 is associated with ER expression.
Materials and Methods
Materials
Fetal calf serum (FCS) (CSL, Melbourne, Australia) and male human serum (HS) (Victorian Red Cross Blood Service) were charcoal-stripped (ch-FCS, ch-HS) for hormone experiments and contained less than 1 pg/ml 17ß-estradiol by RIA. 17ß-estradiol, progesterone, and Cell Dissociation Solution (CDS) were from Sigma-Aldrich (St. Louis, MO). Recombinant human (rh) VEGF165, biotinylated rhVEGF165, biotinylated soybean trypsin inhibitor, RDF buffer and avidin-fluorescein isothiocyanate (avidin-FITC) were purchased as a kit from R&D Systems (Minneapolis, MN). Mouse antihuman CD31 was from DAKO Corp. (Carpinteria, CA), mouse antihuman ER was from Novocastra Laboratories, Ltd. (Newcastle upon Tyne, UK) and rabbit antihuman ERß (catalog no. PAI-313) was from Affinity BioReagents, Inc. (Golden, CO) Phycoerythrin (PE)-labeled sheep anti-IgG (Fab’2 fragments) were from Silenus (Victoria, Australia), whereas horseradish peroxidase (HRP)-conjugated sheep anti-IgG was from Zymed Laboratories, Inc. (San Francisco, CA). Western blotting reagents, Blot-Quickblocker, femto/Tris-buffered saline-Tween 20 (TBST) and femtoLUCENT chemiluminescence detection system were from Chemicon (Temecula, CA). CellTitre 96 MTS [3-4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulphophenyl)-2H-tetrazolium] reagent was from Promega Corp. (Madison, WI). ICI 182,780 was kindly provided by Dr. Frances Sutcliffe (AstraZeneca, Cheshire, UK). VEGFR-1 (flt-1, clone MAFL6) and VEGFR-2 [kinase domain region (KDR), clone MAKD6] were kindly provided by Dr. Napoleon Ferarra from Genentech, Inc. (San Francisco, CA).
Human tissues
Human myometrial tissue was obtained from 35 ovulating women (aged 27–53 yr) who had not taken exogenous hormones in the previous 3 months and underwent hysterectomy for uterine prolapse or fibroids (benign tumor of the myometrium). Informed consent was obtained from each patient, and ethical approval was obtained from Monash Medical Centre Human Research and Ethics Committee B. Many of the hysterectomy samples contained fibroids, which are easily recognized by their well circumscribed border, and were removed before processing. The tissue was collected in HEPES-buffered M199 medium, containing 10% FCS and antibiotic-antimycotic solution (final concentrations: penicillin 106 U/liter, streptomycin 106 U/liter and fungizone 2.5 mg/liter), stored overnight at 4 C, and then processed.
Microvascular EC culture
Myometrial tissue was dissociated with collagenase and deoxyribonuclease, followed by a short trypsin treatment to produce single cell suspensions, and MEC were isolated by positive selection with Ulex europeaus agglutinin-1-coated Dynabeads (Dynal, Oslo, Norway) as described (20). MEC were then seeded into fibronectin-coated tissue culture flasks (10 mg/liter) and cultured in M199 medium containing 15% HS/5% FCS, 2 mmol/liter glutamine, 5 µg/liter bFGF, 100 mg/liter heparin, and antibiotic/antimycotic solution as described (20). On subsequent passages, MEC were repurified with Ulex europeaus agglutinin-1-coated Dynabeads and were used for all analyses between passages 1–3; purity was more than 99% by flow cytometry of CD31-immunolabeled cells. Not all analyses were performed on each isolate, due to insufficient numbers of highly purified MEC obtained from some samples.
Flow cytometric analysis of rhVEGF165 binding to myometrial MEC
Confluent MEC cultures in six-well plates were incubated for 48 h in hormone-free medium (phenol red-free M199 with 15% ch-HS/5% ch-FCS, 10 ng/ml VEGF165, and 100 mg/liter heparin) and then quiesced in HS-free phenol red-free M199 containing 5% ch-FCS, 10 µg/liter VEGF165 and 100 mg/liter heparin for 24 h. We have previously established that myometrial MEC do not proliferate in 5% FCS containing VEGF (20). MEC were then stimulated with 17ß-estradiol, progesterone, or 17ß-estradiol and progesterone in HS-free medium containing 0.5% BSA, 10 µg/liter VEGF165, and 100 mg/liter heparin. In some experiments, 5% ch-FCS was substituted for 0.5% BSA in the HS-free medium. Preliminary experiments showed that it was necessary to include VEGF in human serum-free medium to prevent MEC apoptosis and obtain basal rhVEGF165 binding. Control samples were incubated with vehicle. MEC were harvested in CDS, washed, and resuspended in RDF buffer. Aliquots (25 µl; 5 x 104 cells) were incubated with saturating amounts of biotinylated-rhVEGF165 (30 ng) and anti-CD31 (4 mg/liter) for 1 h at 4 C, followed by 10 µl avidin-FITC for 30 min at 4 C. Cells were washed and then incubated with PE-conjugated antimouse IgG (1/100) for 30 min at 4 C. Nonspecific binding was assessed by substituting biotinylated soybean trypsin inhibitor (same biotin:protein ratio) for biotinylated-rhVEGF165, and CD31 with isotype matched IgG at equivalent concentrations (negative control). Samples were analyzed using a MoFlo flow cytometer (Cytomation, Fort Hills, CO) equipped with an argon laser (Cytomation) and Cyclops Summit software (Cytomation). Two parameter histograms of FITC-VEGF and PE-CD31 were used to determine mean fluorescence intensity (MFI) of VEGF binding on more than 5000 CD31+ MEC per sample. The negative control MFI was subtracted from the respective sample MFI.
Analysis of mRNA for ER and ERß by RT-PCR and Southern blotting
Total RNA was prepared from highly purified cultured MEC (>99%) using the QIAGEN RNeasy Mini Kit. One microgram of total RNA was reverse transcribed and amplified using universal primers for both ER and ERß (sense primer: 5'-CCG GAA TTC TTC/T GAC ATG CTC/G CTGG; antisense primer: 5'-GAT GC/TT CCA TGC CC/TT TGT TAC TC) and for ß2-microglobulin (sense primer: 5'-TGA ATT GCT ATG TGT CTG GGT-3'; antisense primer: 5'-CCT CCA TGA TGC TGC TTA CAT-3') in a single stage PCR for 30 cycles as previously described (21). PCR products were probed with gene-specific 32P-labeled probes (ER probe: 5'-GGT TGT GTG CCT CAA ATC TAT TAT TT; ERß probe: 5'-ATA TCT CTG TGT CAA GGC CAT GA) as described (21).
ER expression by flow cytometry
The presence of ER receptors was determined in MEC suspensions by microwave permeabilization (22) and flow cytometry (23). MEC were harvested with trypsin (0.025%), washed and 106 cells fixed in 2% paraformaldehyde for 30 min at 4 C, washed and resuspended in 0.01 mol/liter citrate buffer, pH 6.0, and microwaved for 30 sec on High (500W), washed and resuspended in PBS/1% FCS. Aliquots (50 µl; 5 x 104 cells) were incubated with 7 mg/liter anti-ER in 10% goat serum for 60 min at 4 C, washed and incubated with PE-antimouse IgG in 10% goat serum for 30 min at 4 C and examined by flow cytometry. The MFI of single parameter histograms of more than 5000 cells was obtained, the MFI of IgG controls subtracted, and the percentage of positive cells with fluorescence intensity more than 98% of control cells determined. For the negative controls, mouse IgG1 (7 mg/liter) was substituted for primary antibody. ER positive (MCF-7 or T47D) and negative (MDA-DB-453) control cells were analyzed in each batch.
Western blot analysis for ER and ERß
MEC lysates (20 µg protein) were denatured in SDS sample buffer, separated by 8% SDS-PAGE and transferred electrophoretically to nitrocellulose membranes (Bio-Rad Laboratories, Inc., Hercules, CA). Membranes were blocked in 5% Blot-QuickBlocker for 30 min at 22 C, incubated for 90 min at 22 C with ER antibody (0.93 mg/liter) in femto/TBST buffer or ERß (1 mg/liter) in femto/TBST/2% Blot-QuickBlocker, followed by incubation for 60 min at 22 C with HRP-antimouse IgG (1/2000 in femto/PBS Tween 20) for ER or HRP-antirabbit IgG (1/3000 in femto/TBST/2% Blot-QuickBlocker) for ERß, and detected by femtoLUCENT chemiluminescence system.
Flow cytometric analysis of VEGFR-1 and VEGFR-2 expression on myometrial MEC
MEC were cultured in T25 flasks, incubated with hormone-free medium, quiesced in human serum-free medium, and incubated with 0 (vehicle), 1, and 10 nmol/liter 17ß-estradiol as described for rhVEGF165 binding. MEC were harvested in CDS and analyzed for rhVEGF165 binding as described, as well as VEGFR-1 and VEGFR-2 expression by flow cytometry. Aliquots (50 µl; 5 x 104 MEC) were incubated with 20 mg/liter anti-VEGFR-1 or 10 mg/liter anti-VEGFR-2 in PBS containing 10% goat serum for 60 min at 4 C, washed, and incubated with PE-antimouse IgG in 10% goat serum for 30 min at 4 C and examined by flow cytometry. The MFI of single parameter histograms of more than 5000 cells was obtained and the MFI of IgG controls subtracted. For negative controls, mouse IgG1 (10 or 20 µg/ml) was substituted for primary antibody. CEM-A7R cells served as negative control cells, and the MFI for both VEGFR-1 or VEGFR-2 antibodies was the same as the IgG control.
MEC proliferation assay
MEC proliferation was assessed using the CellTitre 96 MTS bioassay (Promega Corp.) as previously described (20). Briefly, MEC (4 x 103) were seeded in triplicate into wells of a 96-well plate, made quiescent, and then incubated at 37 C in 5% CO2 with or without 17ß-estradiol (10 nmol/liter) in phenol-red free M199 medium containing 2 µg/liter VEGF and 5% ch-HS/15% ch-FCS in the presence or absence of ICI 182,780 (1 µmol/liter) added 1 h before 17ß-estradiol, with medium changes every 2 d. After 6 d incubation, medium was changed to M199/5% ch-FCS, 20 µl MTS reagent added and incubated for 2 h at 37 C. Absorbance was then measured at 490 nm using a plate reader.
Statistical analysis
Data were analyzed using two-tailed t test or one-way ANOVA, followed by Dunnett’s test for comparison between samples and control using PRISM (GraphPad Software, Inc., San Diego, CA). Correlations were performed using least squares regression analysis and Spearman rank correlation coefficients (RS) were determined using SPSS version 10.0 (SPSS, Inc. Australasia, North Sydney, Australia). The association between estrogen responsiveness and ER expression was examined by Fisher’s exact test using SPSS.
Results
Effect of sex steroids on rhVEGF165 binding to myometrial MEC
The effect of 17ß-estradiol and progesterone on VEGFR expression was examined on confluent, quiescent myometrial MEC in sex steroid hormone-free minimal medium using a ligand binding assay and flow cytometry. Under these conditions, the level of biotinylated-rhVEGF165 binding to myometrial MEC represented basal VEGFR expression and was less than nonconfluent, proliferating MEC (Fig. 1A
). MEC incubated with 17ß-estradiol (10 nmol/liter), but not progesterone (100 nmol/liter), bound more biotinylated-rhVEGF165 than vehicle-treated control cells (basal binding), demonstrated by an increase in fluorescence intensity (Fig. 1A). Figure 1B shows that 10 nmol/liter 17ß-estradiol significantly increased VEGF165 binding (P = 0.01). In contrast, progesterone (100 nmol/liter) alone had no effect on VEGF165 binding and when progesterone was incubated with 17ß-estradiol, there was a nonsignificant reduction of approximately 60% in the 17ß-estradiol-induced increase. Basal rhVEGF165 binding and the 17ß-estradiol effect on rhVEGF165 binding to myometrial MEC were variable between isolates, with a significant proportion that were unresponsive (Table 1). For characterizing the effect of 17ß-estradiol on rhVEGF165 binding, our experiments were limited to 17ß-estradiol-responsive MEC isolates, and we did not further study the effect of progesterone.
|
|
). Incubation times greater than 18 h were not studied as MEC began to undergo apoptosis with prolonged incubation in HS-free conditions. 17ß-estradiol also increased rhVEGF165 binding in a dose-dependent manner, reaching a maximum between 0.1 and 10 nmol/liter (Fig. 2B). The concentration-response curve was biphasic, and at higher 17ß-estradiol concentrations the effect was reduced. The 17ß-estradiol concentration giving maximal rhVEGF165 binding varied between isolates and would peak at 0.1, 1, or 10 nmol/liter 17ß-estradiol for any given MEC isolate. We therefore used two 17ß-estradiol concentrations (0.1 or 1, and 10 nmol/liter) to further examine 17ß-estradiol-stimulated increase in rhVEGF165 binding.
|
and ERß mRNA and protein
To determine whether the 17ß-estradiol effect on rhVEGF165 binding was associated with ER expression, we examined highly purified responsive and nonresponsive MEC (>99% CD31+) by semiquantitative RT-PCR using primers specific for human ER and ERß. Figure 3 shows Southern blots of nine myometrial MEC isolates examined. All expressed ERß as the major transcript, whereas ER was expressed at lower and more variable levels. Figure 3 shows that ER is expressed in 5/9 of the MEC cultures, specifically samples 3–5, 7 and 9. This ER expression was not due to contaminating myometrial smooth muscle cells, which only expressed ER (24) because ER expression was still observed after FACS sorting CD31-labeled MEC (Fig. 3, sample 9). MEC expression of ER and ERß transcripts was observed from passages P1 to P7 in four different MEC isolates (24).
|
and ERß protein in 17ß-estradiol-responsive and -nonresponsive myometrial MEC was then examined by flow cytometry and Western blotting. Figure 4A shows flow cytometric traces from two representative MEC samples, one expressing ER protein shown by a rightward shift in MFI. Low level and variable expression of ER protein was observed in 4/8 samples (Fig. 4, A and B). Western blotting confirmed the variable expression of ER in myometrial MEC isolates, showing a band at approximately 67 kDa, the expected molecular size for human ER protein (Fig. 4C). ERß protein was examined by Western blotting, because highly specific ERß antibodies suitable for flow cytometry are not yet commercially available (25). In all MEC isolates examined, the long form of ERß (26) was present, demonstrated by a band at around 59 kDa (Fig. 4C). There was correspondence in ER expression in the 8/9 MEC samples examined at both mRNA and protein level (Figs. 3
and 4, and data not shown).
|
expression
There was a strong association between ER expression in myometrial MEC and the ability of 17ß-estradiol to increase rhVEGF165 binding. 17ß-estradiol increased rhVEGF165 binding in 9 of the 16 MEC samples, and in 7 of these, ER mRNA and/or protein was also expressed. Of the seven samples that failed to respond to 17ß-estradiol, six did not express ER mRNA and/or protein, and only one was ER positive. Contingency table analysis of this data (Table 1) demonstrates a significant relationship between 17ß-estradiol- mediated increase in rhVEGF165 binding in MEC for both ER mRNA expression (P = 0.03) and ER protein expression (P = 0.025). However, ERß mRNA and/or protein was expressed in all MEC samples examined, whether or not they responded to 17ß-estradiol (Table 1).
17ß-estradiol increases VEGFR-2 expression in ER positive myometrial MEC
rhVEGF165 binds to both VEGFR-1 and VEGFR-2, and we wished to determine whether the increased rhVEGF165 binding was due to 17ß-estradiol up-regulation of one or both receptors. We examined the effect of 17ß-estradiol on VEGFR-1 and VEGFR-2 surface protein expression by flow cytometry using a specific monoclonal antibody to each receptor and compared this expression with 17ß-estradiol- induced rhVEGF165 binding on four ER positive and six ER negative MEC samples. Figure 5A shows representative flow cytometry traces of VEGFR-2 expression in an ER-expressing (ER+) MEC sample and an ER negative (ER-) sample treated for 18 h with vehicle or 1 nmol/liter 17ß-estradiol. A rightward shift of the fluorescence intensity histogram obtained from 5000 MEC was observed for the ER+ sample (left panel) compared with the vehicle control but not in ER- MEC, indicating that 17ß-estradiol increased VEGFR-2 expression in the former but not the latter. Figure 5B shows that 17ß-estradiol (1 and 10 nmol/liter) significantly increased VEGFR-2 expression by approximately 25% compared with vehicle in ER+ MEC. Similarly, 17ß-estradiol significantly increased rhVEGF165 binding by 25%. In contrast, 17ß- estradiol had no stimulatory effect on rhVEGF165 binding or VEGFR-2 expression in ER negative MEC samples (Fig. 5C). Rather, 17ß-estradiol decreased rhVEGF165 binding and VEGFR-2 expression in these ER negative MEC samples. There was a significant correlation between the effect of 17ß-estradiol on rhVEGF165 binding and VEGFR-2 expression in both ER+ and ER- MEC samples (RS = 0.66, P = 0.03, n = 10; and RS = 0.71, P = 0.047, n = 10) for 1 and 10 nmol/liter 17ß-estradiol, respectively. 17ß-estradiol had no effect on VEGFR-1 expression in ER positive or negative MEC samples (results not shown).
|
The role of ER in 17ß-estradiol-mediated increase in rhVEGF165 binding was examined on 17ß-estradiol-responsive MEC isolates using an ER antagonist, ICI 182,780, which blocks the action of both ER and ERß (27). MEC isolates were incubated with a maximal 17ß-estradiol concentration (0.1 or 10 nmol/liter) in the presence and absence of 100-fold molar excess ICI 182,780, and rhVEGF165 binding was measured. Figure 6A shows that ICI 182,780 significantly inhibited 17ß-estradiol-induced VEGF-binding (P < 0.05), suggesting that the increase in rhVEGF165 binding induced by 17ß-estradiol was mediated via ER.
|
+ myometrial MEC grown under suboptimal conditions (VEGF, 2 µg/liter and serum, 5% ch-HS/15% ch-FCS) (20). Figure 6B shows that 10 nmol/liter 17ß-estradiol augmented VEGF-stimulated MEC proliferation, and this effect was also inhibited by 100-fold molar excess ICI 182,780. Discussion
A major finding of the present study was that 17ß- estradiol, but not progesterone, increased VEGFR expression in human myometrial MEC as measured by changes in the level of ligand (rhVEGF165) binding. We found that this response to 17ß-estradiol was highly variable between subjects and was mediated by ER. A second major finding was that MEC isolated and cultured from human myometrium, a classic sex steroid-responsive tissue, consistently expressed ERß, whereas the expression of ER was low and variable between subjects. Thirdly, we found a strong association between ER, but not ERß, expression in myometrial MEC and their ability to respond to 17ß-estradiol by increasing rhVEGF165 binding. Finally, we demonstrated that the increased rhVEGF165 binding correlated with increased VEGFR-2 expression.
17ß-estradiol up-regulation of VEGFR-2 expression in human myometrial MEC
In the present study, we showed that approximately 60% of human myometrial MEC samples responded to 17ß- estradiol with an increase in rhVEGF165 binding. This effect of 17ß-estradiol was both time and concentration dependent. 17ß-estradiol elicited a maximal increase in rhVEGF165 binding that includes physiological plasma concentrations for reproductive age women (0.1–2.2 nmol/liter). The response to 17ß-estradiol concentration was biphasic, and the peak concentration varied between individual MEC isolates. Similarly, Suzuma et al. (28) demonstrated that 17ß-estradiol specifically increased VEGFR-2 mRNA expression in a biphasic manner over a similar time frame in bovine retinal EC. Data from the present study using VEGFR-specific antibodies suggest that 17ß-estradiol up-regulates VEGFR-2 rather than VEGFR-1 expression. 17ß-estradiol enhancement of VEGF-induced MEC proliferation observed in the present study and that of Suzuma et al. (28) may be a direct consequence of an estrogen-mediated increase in VEGFR-2 expression because EC proliferation is dependent on VEGFR-2 rather than VEGFR-1 signaling (29). Alternatively, 17ß- estradiol may have an indirect effect on VEGFR-2 expression by increasing VEGF secretion from EC (28) because VEGF gene transcription is regulated by 17ß-estradiol (30). VEGF up-regulates its own receptor, VEGFR-2 (31), although this response was only observed after 24 h (18, 28, 31). 17ß-estradiol augmentation of bFGF-stimulated EC proliferation is an indirect response, resulting from 17ß-estradiol stimulation of bFGF secretion by EC (32). Together, these studies suggest that 17ß-estradiol enhances angiogenesis by augmenting EC proliferation stimulated by two of the most significant angiogenic growth factors, VEGF and bFGF, although different mechanisms maybe involved (6, 28, 32, 33). The 17ß-estradiol enhancement of VEGFR-2 expression and subsequent proliferative effect mediated by VEGF in myometrial MEC may be important for the expansion of the myometrial microvasculature that undoubtedly occurs during myometrial growth associated with pregnancy and in vascular remodeling associated with placentation. It is uncertain whether 17ß-estradiol promotion of angiogenesis is limited to vascular beds of sex steroid-responsive tissues or occurs throughout the vasculature. This effect of 17ß-estradiol has been observed in a range of EC in vivo and in vitro. For example, 17ß-estradiol augments EC proliferation in human coronary artery EC, umbilical vein EC (6, 7), myometrial MEC, and bovine retinal EC (28). 17ß-estradiol also promotes angiogenesis in bFGF-impregnated Matrigel plugs implanted sc in mice, where dermal MEC sprout, proliferate and migrate into the plug (6, 33).
The cardiovascular protection afforded by estrogen replacement therapy may be to promote an EC proliferative response by up-regulating VEGFR-2 when vessel endothelium is damaged. This effect may not just be confined to large vessels but may contribute to protection of endothelium in other tissues, sensitizing EC to low levels of angiogenic growth factors by up-regulating VEGFR-2. In fact, topical 17ß-estradiol promotes wound healing in the aged, in whom it is typically delayed and in whom circulating 17ß-estradiol levels are low (34). The lack of response by myometrial MEC to progesterone, together with the diminished response observed with 17ß-estradiol and progesterone, may explain why the inclusion of progesterone in hormone replacement therapy is not as beneficial as a cardiovascular protective agent as estrogen alone (35).
ER and ERß expression in human myometrial MEC
The second major finding of this study was that cultured myometrial MEC constitutively express ERß, whereas ER expression was variable between subjects. This is the first study that has examined ER and ERß expression in purified human MEC in a cohort of subjects large enough to detect the level of variability observed for ER expression. We found that ER (when present) and ERß mRNA expression was consistently demonstrated for up to seven passages. Others have demonstrated ER in human large vessel and endometrial MEC, but the specific type of ER was not determined (7, 9, 10). In an immunohistochemical study of human tissues, the great vessels were both ER and ERß positive (36), whereas others found cultured umbilical vein EC expressed ERß, but not ER mRNA (37, 38). However, in immunohistochemical studies of myometrial tissue, ERß but not ER was demonstrated in rat EC (39), but in mouse and rhesus monkey, both ER and ERß were expressed (40, 41). Together, these studies indicate that ERß expression may be constitutive in large and microvessel EC and that ER expression is variable and may depend on EC phenotype and the conditions under which they were evaluated.
Role of ER and ERß in 17ß-estradiol up-regulation of VEGFR-2 in human myometrial MEC
A major finding of the present study was the significant association between ER expression and the ability of myometrial MEC to functionally respond to 17ß-estradiol by increasing rhVEGF165 binding and VEGFR-2 expression. While our data on ER expression is only semiquantitative, it is striking that in general those samples expressing ER responded to 17ß-estradiol, whereas those lacking ER failed to respond. This, together with the time taken (several hours) for 17ß-estradiol to elicit this response, and its inhibition by the antiestrogen, ICI 182,780, provides three lines of evidence suggesting that the 17ß-estradiol-mediated increase in VEGFR-2 expression is genomic and mediated by ER, most likely ER. It is uncertain whether the 17ß-estradiol-induced increase in VEGFR-2 expression in bovine retinal EC was mediated by ER, because the effect of an ER antagonist was not examined (28). It is possible that 17ß-estradiol via ER directly regulates VEGFR-2 expression, because the VEGFR-2 promoter contains 5 Sp1 binding sites (42) and is regulated by the Sp1 transcription factor (43). Furthermore, 17ß-estradiol activates gene transcription through interaction with Sp1 proteins via ER, but not ERß (44).
The variable expression of ER in myometrial MEC, as well as the variability in functional response to 17ß-estradiol, suggests that other factors may influence these parameters. However, we have found no relationship between age of patient, indication for hysterectomy (mainly fibroids), stage of menstrual cycle when hysterectomy was performed, or the ability of MEC to grow well in culture, and ER expression. Although MEC were obtained from the inner two thirds of the myometrium (20), because the inner layers show greater sex steroid responsiveness (45), it is possible that sampling different sites of the uterine wall may influence ER expression. It will be important to determine whether similar variability in ER expression occurs in other vascular beds and in particular in the large vessels commonly affected by atherosclerosis. Similar variability in ER expression was observed in a study of cultured vascular smooth muscle cells (VSMC) from human coronary arteries, where 53% of subjects expressed ER (46). While this study could not examine for ER and ERß expression, the authors did find a significant association between VSMC expression of ER and the absence of atherosclerosis in premenopausal women. These data, together with our own, raise interesting clinical questions of whether such a correlation exists between MEC expression of ER in our cohort of patients, and whether the vascular response of these women to estrogen replacement therapy would also vary. With respect to the myometrial microvasculature, variable ER expression may result in the variable degrees of uterine atrophy observed in women taking hormone replacement therapy, because the microvasculature may regulate tissue mass (47).
We have demonstrated a significant association between ER expression and the functional response mediated by 17ß-estradiol, but definitive proof awaits the use of ligands or antagonists with sufficient specificity for ER and ERß. Much research is being devoted to the identification of such specific ligands; however, those currently available are not of sufficient specificity (48) for testing in MEC coexpressing both ER and ERß. Because ER and ERß can heterodimerize (49), and ERß can operate as a negative regulator of target gene transcription in a promoter-dependent manner (25, 49), the ratio of ERß:ER may be more important in determining the final outcome of 17ß-estradiol effects on myometrial MEC and EC in general. In human VSMC, where ERß predominates, the ratio of ERß:ER varied considerably between individuals (25). It remains to be determined whether the ERß:ER ratio is similar for EC and VSMC in a given vascular bed for an individual and whether the function of target genes in these two vascular cell types are altered similarly. Differences already exist, because 17ß-estradiol has opposing activities on EC and VSMC proliferation (2, 6, 7, 28, 41, 50). Studies from ER knockout mice suggest that 17ß-estradiol-dependent angiogenesis and re-endothelialization involve ER (33, 51), but that ERß is important in vascular response to injury, which involved both VSMC and EC (50, 52). It is likely that the cardiovascular protective effect of estrogen results from multiple effects of 17ß-estradiol on different cellular components of the blood vessel wall, which in turn are dependent on complex interactions between ER and ERß, the response elements present in target genes and the complement of coregulators in both cell types. Although 17ß-estradiol may have different functional responses on EC and VSMC, these actions may together contribute to the cardiovascular protection afforded by 17ß-estradiol. The level of protection may be quite different between individuals, depending on the level of endothelial ER expression, which is expected to affect uterine MEC and may involve the entire vasculature.
Acknowledgments
We thank the gynecologists at Monash Medical Centre for provision of hysterectomy tissue, Nancy Taylor for collection of tissue, Mal Forsyth and the Australian Red Cross Blood Service, Melbourne for the provision of male human serum, and Paul Hutchinson, Department of Immunology, for assistance with the flow cytometry.
Footnotes
This study was supported by the Australian National Health and Medical Research Foundation Grant 124331 (to P.A.W.R.).
Abbreviations: bFGF, Basic fibroblast growth factor; CDS, Cell Dissociation Solution; ch, charcoal stripped; EC, endothelial cell(s); ER, estrogen receptor; FCS, fetal calf serum; FITC, fluorescein isothiocyanate; HRP, horseradish peroxidase; HS, human serum; MEC, microvascular endothelial cell(s); MFI, mean fluorescence intensity; MTS, 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4sulpho-phenyl)-2H-tetrazolium; PE, phycoerythrin; PR, progesterone receptor; rh, recombinant human; RS, Spearman rank correlation coefficients; TBST, Tris-buffered saline-Tween 20; VEGF, vascular endothelial growth factor; VEGFR, VEGF receptor; VSMC, vascular smooth muscle cells.
Received April 9, 2001.
Accepted May 28, 2002.
References
mediates the nongenomic activation of endothelial nitric oxide synthase by estrogen. J Clin Invest 103:401–406[Medline]
and ß: two receptors of a kind? Curr Med Chem 7:561–576[Medline]
and ERß at AP1 sites. Science 277:1508–1510 and ERß. Proceedings of the 10th International Congress of Endocrinology, Sydney, Australia, 2000, p 377 (Abstract P1128)
and ß. Prevalence of estrogen receptor ß mRNA in human vascular smooth muscle and transcriptional effects. Circulation 101:1792–1798 in vivo and in vitro. Biochem Biophys Res Commun 243:122–126[CrossRef][Medline]
and ß. Proc Natl Acad Sci USA 97:10972–10977 and ß in the rat uterus. Biol Reprod 61:955–964 and ER ß mRNA in intrauterine tissues of the pregnant rhesus monkey. Am J Physiol 278:C190–C198
/ß)-dependent activation at GC-rich (Sp1) promoter elements. J Biol Chem 275:5379–5387 and estrogen receptor ß to partial estrogen agonists/antagonists. Mol Pharmacol 54:105–112 and ß form heterodimers on DNA. J Biol Chem 272:19858–19862-deficient mice. Nat Med 3: 545–548
but not estrogen receptor-ß. Circulation 103:423–428This article has been cited by other articles:
 |
|
 |
|
  D. G.M. Molin, N. M.S van den Akker, and M. J. Post Affirmative Action of Osteopontin on Endothelial Progenitors Arterioscler Thromb Vasc Biol, December 1, 2008; 28(12): 2099 - 2100. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zaitseva, B. J. Vollenhoven, and P. A.W. Rogers Retinoids regulate genes involved in retinoic acid synthesis and transport in human myometrial and fibroid smooth muscle cells Hum. Reprod., May 1, 2008; 23(5): 1076 - 1086. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zaitseva, B. J. Vollenhoven, and P. A.W. Rogers In vitro culture significantly alters gene expression profiles and reduces differences between myometrial and fibroid smooth muscle cells Mol. Hum. Reprod., March 1, 2006; 12(3): 187 - 207. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M A J Herve, G Meduri, F G Petit, T S Domet, G Lazennec, S Mourah, and M Perrot-Applanat Regulation of the vascular endothelial growth factor (VEGF) receptor Flk-1/KDR by estradiol through VEGF in uterus J. Endocrinol., January 1, 2006; 188(1): 91 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Hussein-Fikret, P. J. Fuller, and C. E. Gargett Expression of Steroid Receptor Coactivators in Cultured Cells From Paired Myometrial and Fibroid Tissues Reproductive Sciences, September 1, 2005; 12(6): 445 - 451. [Abstract] [PDF]
|
|
|
|
|
|
L. C. Ajonuma, F. O. Ajuonuma, and C. L. Chukwu Coasting acts through downregulation of VEGF gene expression and protein secretion Hum. Reprod., February 1, 2005; 20(2): 574 - 575. [Full Text] [PDF]
|
|
|
|
|
|
M. Zaitseva, D. S. Yue, J. A. Katzenellenbogen, P. A. W. Rogers, and C. E. Gargett Estrogen Receptor-{alpha} Agonists Promote Angiogenesis in Human Myometrial Microvascular Endothelial Cells Reproductive Sciences, December 1, 2004; 11(8): 529 - 535. [Abstract] [PDF]
|
|
|
|
|
|
J.A. Garcia-Velasco, A. Zuniga, A. Pacheco, R. Gomez, C. Simon, J. Remohi, and A. Pellicer Coasting acts through downregulation of VEGF gene expression and protein secretion Hum. Reprod., July 1, 2004; 19(7): 1530 - 1538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Elkin, A. Orgel, and H. K. Kleinman An Angiogenic Switch in Breast Cancer Involves Estrogen and Soluble Vascular Endothelial Growth Factor Receptor 1 J Natl Cancer Inst, June 2, 2004; 96(11): 875 - 878. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Dietrich, A. Haitel, J. C. Huber, and W. J. Reiter Expression of Estrogen Receptors in Human Corpus Cavernosum and Male Urethra J. Histochem. Cytochem., March 1, 2004; 52(3): 355 - 360. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and VEGFR-2 expression (
) then measured by flow cytometry. A, Representative flow cytometry histograms of VEGFR-2-FITC fluorescence for ER