A high level of estrogen receptor-α (ER-α) is believed to be favorable in the prognosis and treatment of certain female cancers. ER-α expression in the ER-negative breast cancer cell lines inhibits their proliferation and invasive, metastatic potential in vitro. We stably overexpressed the ER-α in the human endometrial cancer cell line Ishikawa and showed that, unlike estradiol, high levels of ER-α significantly inhibit the growth of tumors xenografted from the Ishikawa cells. Subsequent to ER-αoverexpression, in vivo down-regulation of vascular endothelial growth factor was observed in tumor xenografts. In addition, these tumors showed an inhibition of vascularization and of the angiogenic agent, integrin αvβ3. Involvement of a switch in the angiogenic pathways during tumorigenesis has been a recent focus of interest. Our results indicate that a high level of ER-α may be beneficial in the control of female cancers because of its inhibitory effect on such angiogenic pathways.

A high level of ER-α2 is believed to be favorable in the control of ovarian (1) and endometrial (2) cancers. In the past, we (10) and others have reported the development of cell lines that were stably transfected with ER-α and were used as models to study the mechanisms involved in the estrogen/ER-mediated control of cellular proliferation(3). In most cases, these cell lines have shown inhibition in their growth (3) and metastatic/invasive potential(4), possibly through an effect of ER-α on certain growth-regulating genes, such as c-myc(5).

In the ER-α-overexpressing clone of Ishikawa cell line (ISH-ER)discussed here, we previously reported an ER-α-mediated inhibition of in vitro growth.3 Accompanying this growth inhibition were a stimulation of NOS activity and VEGF levels,3 both of which are critical in determining vascular physiology and mitotic behavior. Involvement of a switch in the angiogenic pathways during tumorigenesis has been a recent focus of interest (6). The process of angiogenesis determines the available supply of blood to a growing cancer and therefore plays an important role in its progression. Vasoactive factors such as NOS and VEGF may therefore be involved in the mechanism by which estrogen influences cancer growth. In this study, we investigate this aspect in vivo. We report an ER-α-mediated in vivo growth inhibition of tumors xenografted from parent and ER-transfected Ishikawa cells. In addition,we demonstrate an inhibition of VEGF and other angiogenic parameters after the ER-α overexpression in the Ishikawa cells. The present work indicates that a high level of ER-α may be beneficial in controlling female cancers because of its inhibitory effect on angiogenic pathways.

Athymic Mice.

Three-week-old nude female mice [strain Tac:Cr:(NCr)-nufBR] were purchased from Taconic (Germantown, NY). The animals were housed in an aseptic environment under controlled conditions of light and humidity and received food and water ad libitum. Animals were allowed to acclimate to the new environment for 1 week before ovariectomy or sham operation. For the week after the surgery, animals were kept on tetracycline water. The success of ovariectomy was ascertained by microscopic examination of vaginal swabs confirming that the mice were in a diestrous cycle. The blood estrogen levels in the OVX, SHAM, and intact animals were confirmed by radioimmunoassay.

Inoculation of Athymic Mice.

Cells were routinely grown in Medium 199 (Sigma, St. Louis, MO)supplemented with 10% fetal bovine serum (Hyclone, Logan, UT). On the day of inoculation, cells were trypsinized and resuspended in Matrigel(Fisher Scientific Co., Pittsburgh, PA) at 1 million cells per 100μl. Animals received inoculations containing 100 μl of this suspension in each flank. Tumors were allowed to grow, and their dimensions were measured weekly. Tumor weight was determined as follows(7, 8):

\[\mathrm{Estimated\ tumor\ weight\ (mg)}{=}\ \frac{{\{}\mathrm{Longest\ diameter\ (mm)}{\times}{[}\mathrm{shortest\ diameter\ (mm)}{]}^{2}{\}}}{2}\]

Treatment.

One week after the ovariectomy, animals were inoculated with the cell suspension and implanted with 60-day time-release pellets of E2, TAM,OHT, or a placebo (Innovative Research of America, Sarasota,FL). OHT was obtained from Sigma and was custom-prepared into 60-day time-release pellets at Innovative Research of America. A 12-gauge trochar was used to insert the pellet s.c. at the back of each animal’s neck. Sixty days later, if an animal was still under observation, it was implanted with a fresh pellet.

The animal protocols described above were approved by the Animal Research Committee at the State University of New York at Buffalo. All experiments on the animals were performed according to the NIH-endorsed guidelines.

Immunostaining for VEGF and Integrinα vβ3.

At the end of each experiment, the animals growing tumors were euthanized, and their tumors were excised and preserved in formalin until use. The tumors were embedded in paraffin, sectioned, and stained for either integrinα vβ3 (using the anti-αvβ3 antibody LM609; R&D Systems, Inc., Minneapolis, MN) or for VEGF (using an anti-VEGF antibody; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). The stained slides were then scored for the intensity of stain against the background of their respective negative controls stained in a similar fashion, except that the antibody was replaced with pre-immune serum. The intensity of stain thus measured was expressed as the level of the VEGF or the integrinα vβ3 in tumors.

Staining for Vascularization.

The slides were prepared as for the VEGF andα vβ3 staining. This staining was performed using the Verhoeff elastic stain method(9). Briefly, the slides were stained with a hematoxylin-ferric chloride iodine solution, which makes the vascular elastic fibers and the cell nuclei appear black. The slides were then scored for the magnitude of black-stained fibers, which was expressed as the level of vascularization in tumors.

In Vitro Assays.

The proliferation assays were performed as described by Ali et al.(10). Briefly, equal numbers of parent,ER-transfected, and control cells were seeded in Dulbecco’s Minimal Medium containing 10% fetal bovine serum. The cells were then counted after trypsinization on alternate days over a period of 10 days. VEGF was also measured as described by Ali et al.(10). Cells were seeded in equal numbers, and VEGF was measured in the spent medium after 48 h with the Quantikine VEGF immunoassay (R&D Systems). To assay NOS activity, the cells were grown in the same way. NOS activity was measured in cell lysates with the NOSdetect kit (Stratagene). This kit measures the activity of NOS in an NOS-catalyzed coupled reaction where citrulline and nitric oxide are produced from arginine in equimolar amounts. The amount of citrulline thus produced is measured and expressed in terms of the activity of NOS. The final values for NOS activity were corrected for total cell protein in each sample.

Effect of ER Overexpression on Tumor Growth.

As described elsewhere,3 to develop the cell line stably overexpressing the ER-α, Ishikawa cells were cotransfected with a G418 resistance-conferring plasmid and an ER-α-expressing vector. The ER-overexpressing clones were selected for G418 resistance and then screened for high ER-α expression. All of the subsequent experiments, including the ones presented here, were then performed using the parent Ishikawa cells, an ER-overexpressing clone of Ishikawa(ISH-ER), and a control-transfected clone of Ishikawa (ISH-NON). ISH-NON was a G418-resistant clone that was selected during the screening for ISH-ER. Having gone through the same transfection process, ISH-NON was deemed a logical control against the phenotypic changes arising from the transfection protocol alone. In addition,ISH-NON was also a control against clone-dependent artifacts.

To establish a baseline, each of the cell lines, i.e.,Ishikawa, ISH-ER, and ISH-NON, was inoculated into the flanks of OVX or SHAM nude female mice. No visible tumors formed from any of the cell lines in either SHAM or OVX animals observed over a period of 5 weeks(Fig. 1 A).

OVX animals inoculated with Ishikawa, ISH-ER, and ISH-NON cells were implanted with 0.15-mg E2 pellets, which were formulated to produce a blood E2 level of 50–75 pg/ml (3-fold higher than the physiological E2 level in the nude mouse strain used). This group formed palpable tumors within 3 weeks (Fig. 1, A and C). Tumor growth was more aggressive in mice inoculated with parent Ishikawa and control ISH-NON cells, whereas it was significantly slower in the case of ER-transfected ISH-ER cells. In mice that had received implants of 1.5-mg E2 pellets (producing blood E2 levels of >900 pg/ml), an insignificant increase in tumor size was observed in the case of Ishikawa, ISH-NON, and ISH-ER cells. In all groups, therefore, E2 at levels higher than physiological stimulated tumor growth (Fig. 1, A and C). In contrast, ER was found to inhibit the growth of these tumors.

When OVX mice received implants containing 1.5-mg TAM or OHT pellets(producing blood levels of 3–4 ng/ml in each case), palpable tumors from Ishikawa cells developed within 3 weeks. ISH-ER cells,however, did not form any tumors (Fig. 1 B). Moreover, the tumors formed in these groups were significantly smaller than the tumors in the corresponding groups that had received E2 pellets. This indicated that the estrogenic effect required for tumor formation was weaker in the case of antiestrogens compared with E2. In these groups,therefore, antiestrogens were found to be stimulatory to tumor growth,but to a lesser extent than E2. ER overexpression was found to consistently inhibit cell growth.

Effects on Integrin αvβ3, VEGF, and Vascularization in Tumors.

To examine the levels of VEGF andα vβ3, immunostaining with an anti-VEGF or an anti-αvβ3 antibody,respectively, was performed on sections from tumors initiated in nude mice with parent and ER-transfected Ishikawa cells. Slides from these tumors were also stained by a hematoxylin-ferric chloride-iodine solution to examine the extent of vascularization. These tumors were excised from the animals that had received implants of 0.15-mg E2 pellets. The tumors initiated from the ER-transfected cells showed a significant down-regulation of VEGF (Fig. 2) and integrinα vβ3 (Fig. 3). As anticipated, the magnitude of vascularization was also found to be decreased in the ER-overexpressing tumors (Fig. 4).

In the tumors initiated from Ishikawa cells, therefore, ER overexpression was found to inhibit the expression of integrinα vβ3 and VEGF and the degree of vascularization.

ER-α overexpression in Ishikawa cells was found to be inhibitory to in vivo tumor growth and to the angiogenic parameters studied here. We previously reported the development of this ER-α-overexpressing clone of Ishikawa cell line, ISH-ER, that has an ER level 6-fold higher than the parent Ishikawa cells3 (∼50 fmol/mg of WCP in Ishikawa and∼300 fmol/mg of WCP in ISH-ER cells, as determined in a hormone-binding assay). In the previously reported in vitroassays, each of the three cell lines, i.e., the parent Ishikawa, the ER-transfected ISH-ER, and the control-transfected ISH-NON, showed an increase in growth rate when induced with 10−9mE2.3 This was expected of a cell line of endometrial origin. With or without E2 induction, however, the growth rate of ISH-ER was inhibited significantly compared with the parent or control-transfected cells. This indicated that although E2 stimulated the growth of Ishikawa cells, ER inhibited their growth. This growth inhibition was consistently observed in other ER-α-overexpressing clones as well (Table 1), indicating that this effect was not clone-dependent.

In the in vivo experiments reported here, a similar pattern was observed in various ER-overexpressing clones of Ishikawa. Furthermore, in our preliminary work, this pattern was found to be consistent through various strains of nude mice (data not shown). In addition, in our previous work with a human endothelial cell line ECV304 (10), we found a similar growth-inhibitory effect of ER-α overexpression that was also found to be associated with a modulation of certain vasoactive/angiogenic factors. The effects of ER-α overexpression were therefore found to be similar in two cell lines of completely different origins.

As shown in Fig. 1, B and C, in the OVX or SHAM animals, no growth was observed from any of the cell lines tested. When these animals were administered E2, however, each of the three cell lines, Ishikawa, ISH-ER, and ISH-NON, formed tumors. The tumors formed by ISH-ER were, however, significantly smaller than the parent or control cell line. These results clearly indicated that in these animals, the presence of a higher than the physiological level of E2 was required for tumor growth; E2 was therefore stimulatory for the growth of these cells in vivo as well. Consistent with the in vitro results, however, ER was once again found to inhibit the growth of xenografted tumors. A dissociation in the effects of ER and its ligand was therefore consistently observed, indicating a dominant ligand-independent effect of ER after its overexpression. It is possible that this effect is caused by binding of the overexpressed ER to ligands other than E2 (11, 12) or by binding of ER to DNA sites in an unliganded conformation (13).

The effects of E2 and ER-α overexpression observed in the Ishikawa cells are physiologically consistent. In a physiological milieu,endometrial cells are, indeed, stimulated by E2. A high level of ER-α, on the other hand, is known to be favorable for controlling the growth of endometrial and other female cancers (1, 2).

The fact that no tumor growth was observed in SHAM animals indicates that for in vivo growth, Ishikawa cells required a higher level of E2 than the normal physiological level in the nude mouse. Because Ishikawa cells are of human origin, it is quite possible that the E2 blood levels in mice were not high enough to elicit an estrogenic effect in these cells.

The tumors initiated from ER-overexpressing ISH-ER cells showed a significant down-regulation of VEGF compared with the tumors from parent Ishikawa cells (Fig. 2). We previously showed a similar in vitro down-regulation of VEGF levels in cultured ER-overexpressing cells (Ref. 10 and Table 1). Our in vitro and in vivo results, therefore, consistently showed inhibition of VEGF as a result of higher ER-α expression. The effect of estrogen on endometrial growth and endometrial VEGF (14)is known to be stimulatory. Our results, however, show that the effect of ER on these two parameters in a cell line of endometrial origin contrasts with E2, i.e., whereas the hormone is stimulatory,its receptor is inhibitory.

In case of endometrial cancer, ER-positive tumors exhibit an ER level of 155–209 fmol/mg of WCP in a hormone-binding assay(15). As reported previously,3 using a similar technique, we determined that the levels of ER in our Ishikawa (and control ISH-NON) and ER-overexpressing ISH-ER cells were∼50 and ∼300 fmol/mg of WCP, respectively. Our models, therefore,cover a broad physiological range of ER expression found in endometrial cancers.

It has been shown that in women with ovarian (1) or endometrial (2, 16) cancers, a positive ER status is predictive of survival. Our results demonstrating a negative correlation between the growth and ER content of endometrial tumor xenografts support these earlier findings. We show a possible involvement of angiogenic factors in ER-related inhibition of tumor growth. Our in vitro studies with ER-overexpressing clones of Ishikawa have shown a modulation of growth and of angiogenic factors, including VEGF (Table 1). VEGF is one of the key determinants of the onset and progression of angiogenesis. It is possible that a positive correlation between ER levels and the survival rate of ovarian and endometrial cancer patients involves an ER-mediated, VEGF associated negative control of angiogenesis in these cancers. In the present study, the ER-overexpressing tumors that showed down-regulation of VEGF were indeed found to have a significantly lower level of vascularization compared with larger tumors expressing basal levels of ER (Fig. 4).

In the ER-mediated inhibition of angiogenesis observed here, it is possible that other cellular factors that are controlled by angiogenic cytokines (such as VEGF) are influenced in the downstream pathway,eventually leading to the inhibition of vascularization. To investigate this issue, we also examined the effect of ER overexpression on the expression of integrinα vβ3, one of the adhesion molecules that promotes angiogenesis in certain cancers(17) and is positively affected by VEGF (18). We found that the level of integrinα vβ3, as anticipated,was significantly lower in the tumors initiated from ER-overexpressing cells (Fig. 3).

In conclusion, we show here that, in contrast with its ligand E2, high levels of ER-α down-regulate in vivo growth of the endometrial cancer cell line Ishikawa, and that this growth inhibition is associated with a down-regulation of the angiogenic factors VEGF and integrin αvβ3, leading to an inhibition of tumor angiogenesis.

Fig. 1.

A and B, effect of E2 and antiestrogens on in vivo growth of tumors initiated from parent (open columns), ER-transfected(filled columns) or control-transfected (hatched columns) Ishikawa cells. A, 3-week old female nude mice, strain Tac:Cr:(NCr)-nufBR, received implants of 1 × 106 cells in each flank. Tumors were allowed to grow, and their weights were determined weekly. Animals were either OVX or SHAM. Some groups of the OVX animals received s.c. implants of time-release pellets containing 0.15 mg or 1.5 mg of E2. The results are expressed as means; bars, SE(n ≥ 8). B, groups of OVX nude animals also received implants of 1.5-mg time-release pellets of TAM, OHT, or a placebo. The results are expressed as mean; bars, SE (n ≥ 8). C, representative animals at week 5, implanted with 0.15-mg E2 pellets and xenografted with ER-transfected(left) or parent (middle) Ishikawa cells. Dissection of the larger tumor (right) revealed a solid mass devoid of any necrosis or water retention. The growing tumor xenografts can be seen in the flank areas of the animals.

Fig. 1.

A and B, effect of E2 and antiestrogens on in vivo growth of tumors initiated from parent (open columns), ER-transfected(filled columns) or control-transfected (hatched columns) Ishikawa cells. A, 3-week old female nude mice, strain Tac:Cr:(NCr)-nufBR, received implants of 1 × 106 cells in each flank. Tumors were allowed to grow, and their weights were determined weekly. Animals were either OVX or SHAM. Some groups of the OVX animals received s.c. implants of time-release pellets containing 0.15 mg or 1.5 mg of E2. The results are expressed as means; bars, SE(n ≥ 8). B, groups of OVX nude animals also received implants of 1.5-mg time-release pellets of TAM, OHT, or a placebo. The results are expressed as mean; bars, SE (n ≥ 8). C, representative animals at week 5, implanted with 0.15-mg E2 pellets and xenografted with ER-transfected(left) or parent (middle) Ishikawa cells. Dissection of the larger tumor (right) revealed a solid mass devoid of any necrosis or water retention. The growing tumor xenografts can be seen in the flank areas of the animals.

Close modal
Fig. 2.

A, levels of VEGF in tumors initiated from parent (Ishikawa; open column) and ER-transfected(ISH-ER; filled column) Ishikawa cells. The results[mean ± SE (bars); n ≥ 8] are expressed as the VEGF level in the tumors. B, representative slides of a positive control (human tonsil tissue; left) and tumors initiated from ER-transfected (middle) and parent(right) Ishikawa cells. Formalin-preserved tumor samples were embedded in paraffin, sectioned, and stained with an anti-VEGF antibody. The slides were scored for the intensity of positive stain against a background of their respective negative controls, which were stained similarly except that the anti-VEGF antibody was replaced with pre-immune serum. The size of the field of vision and the magnification were same for all slides. The VEGF-positive areas on the slides appear brown against a blue background.

Fig. 2.

A, levels of VEGF in tumors initiated from parent (Ishikawa; open column) and ER-transfected(ISH-ER; filled column) Ishikawa cells. The results[mean ± SE (bars); n ≥ 8] are expressed as the VEGF level in the tumors. B, representative slides of a positive control (human tonsil tissue; left) and tumors initiated from ER-transfected (middle) and parent(right) Ishikawa cells. Formalin-preserved tumor samples were embedded in paraffin, sectioned, and stained with an anti-VEGF antibody. The slides were scored for the intensity of positive stain against a background of their respective negative controls, which were stained similarly except that the anti-VEGF antibody was replaced with pre-immune serum. The size of the field of vision and the magnification were same for all slides. The VEGF-positive areas on the slides appear brown against a blue background.

Close modal
Fig. 3.

A, levels of integrinα vβ3 in tumors initiated from parent(Ishikawa; open column) and ER-transfected (ISH-ER; filled column) Ishikawa cells. The results[mean ± SE (bars); n ≥ 8] are expressed as theα vβ3 levels in tumors. B,representative slides of a positive control (human kidney tissue; left) and tumors initiated from ER-transfected(middle) and parent (right) Ishikawa cells. The slides were prepared as in Fig. 2 and stained using the anti-αvβ3 antibody LM609. The slides were then scored as in Fig. 2 using similar negative controls. The size of the field of vision and the magnification were same for all slides. Theα vβ3-positive areas on the slides appear brown against a blue background.

Fig. 3.

A, levels of integrinα vβ3 in tumors initiated from parent(Ishikawa; open column) and ER-transfected (ISH-ER; filled column) Ishikawa cells. The results[mean ± SE (bars); n ≥ 8] are expressed as theα vβ3 levels in tumors. B,representative slides of a positive control (human kidney tissue; left) and tumors initiated from ER-transfected(middle) and parent (right) Ishikawa cells. The slides were prepared as in Fig. 2 and stained using the anti-αvβ3 antibody LM609. The slides were then scored as in Fig. 2 using similar negative controls. The size of the field of vision and the magnification were same for all slides. Theα vβ3-positive areas on the slides appear brown against a blue background.

Close modal
Fig. 4.

A, vascularization in tumors initiated from parent (Ishikawa; open column) and ER-transfected(ISH-ER; filled column) Ishikawa cells. The results[mean ± SE (bars); n ≥ 8] are expressed as the level of vascularization in the tumors. B, representative slides of a positive control (human aorta tissue; left) and tumors initiated from ER-transfected (middle) and parent(right) Ishikawa cells. The slides were prepared as in Fig. 2, stained using the Verhoeff elastic stain method, and scored for the magnitude of black-stained fibers. The size of the field of vision and the magnification were same for all slides. The extent of vascularization in the slides is observed as the black-stained elastic fibers.

Fig. 4.

A, vascularization in tumors initiated from parent (Ishikawa; open column) and ER-transfected(ISH-ER; filled column) Ishikawa cells. The results[mean ± SE (bars); n ≥ 8] are expressed as the level of vascularization in the tumors. B, representative slides of a positive control (human aorta tissue; left) and tumors initiated from ER-transfected (middle) and parent(right) Ishikawa cells. The slides were prepared as in Fig. 2, stained using the Verhoeff elastic stain method, and scored for the magnitude of black-stained fibers. The size of the field of vision and the magnification were same for all slides. The extent of vascularization in the slides is observed as the black-stained elastic fibers.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

2

The abbreviations used are: ER, estrogen receptor; NOS, nitric oxide synthase; VEGF, vascular endothelial growth factor; OVX, ovariectomized; SHAM, sham-operated; E2, 17-β-estradiol,TAM, tamoxifen; OHT, 4-hydroxytamoxifen; WCP, whole-cell protein.

3

S. H. Ali, A. L. O’Donnell, and P. Dandona. Overexpression of estrogen receptor-α in the endometrial carcinoma cell line Ishikawa: inhibition of growth and vascular endothelial growth factor, stimulation of nitric oxide synthase activity, manuscript in preparation.

Table 1

In vitro effects of ER-α overexpression in clones of Ishikawa cells.

Subsequent to ER overexpression, changes in the rate of cell proliferation, NOS activity, and VEGF levels were studied in cultured cells. The values given are the ranges of the observed parameters,which are expressed as a percentage of the corresponding value in the parent cells. Number of ER-overexpressing clones exhibiting the studied effect is indicated.

Parameter studiedChange observed (%)Number of clones exhibiting the effect
Cell proliferation 22–33 
NOS activity 470–700 
VEGF level 33–50 
Parameter studiedChange observed (%)Number of clones exhibiting the effect
Cell proliferation 22–33 
NOS activity 470–700 
VEGF level 33–50 
1
Geisler J. P., Wiemann M. C., Miller G. A., Zhou Z., Geisler H. E. Estrogen and progesterone receptors in malignant mixed mesodermal tumors of the ovary.
J. Surg. Oncol.
,
59
:
45
1995
.
2
Martin J. D., Hahnel R., McCartney A. J., Woodings T. L. The effect of estrogen receptor status on survival in patients with endometrial cancer.
Am. J. Obstet. Gynecol.
,
147
:
322
1983
.
3
Levenson A. S., Jordan V. C. Transfection of human estrogen receptor (ER) cDNA into ER-negative mammalian cell lines.
J. Steroid Biochem. Mol. Biol.
,
51
:
229
1994
.
4
Garcia M., Derocq D., Freiss G., Rochefort H. Activation of estrogen receptor transfected into a receptor-negative breast cancer cell line decreases the metastatic and invasive potential of the cells.
Proc. Natl. Acad. Sci. USA
,
89
:
11538
1992
.
5
Maminta M. L. D., Molteni A., Rosen S. T. Stable expression of the human estrogen receptor in HeLa cells by infection: effect of estrogen on cell proliferation and c-myc expression.
Mol. Cell. Endocrinol.
,
78
:
61
1991
.
6
Hanahan D., Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis.
Cell
,
86
:
353
1996
.
7
Osborne C. K., Hobbs K., Clark G. M. Effect of estrogens and antiestrogens on growth of human breast cancer cells in athymic nude mice.
Cancer Res.
,
45
:
584
1985
.
8
Kubota T., Josui K., Fukutomi T., Kitajima M. Growth regulation by estradiol, progesterone and recombinant human epidermal growth factor of human breast carcinoma xenografts grown serially in nude mice.
Anticancer Res.
,
15
:
1275
1995
.
9
Carson F. L. Histotechnology: A Self-Instructional Text
147
-149, ASCP Press Chicago  
1990
.
10
Ali S. H., O’Donnell A. L., Mohamed S., Mousa S., Dandona P. Stable overexpression of estrogen receptor-α in ECV304 cells inhibits proliferation and levels of secreted endothelin-1 and vascular endothelial growth factor.
Mol. Cell. Endocrinol.
,
152
:
1
-9,  
1999
.
11
Zava D. T., McGuire W. L. , Androgen action through estrogen receptor in a human breast cancer cell line.
Endocrinology
,
100
:
624
1978
.
12
Zhu B. T., Conney A. H. Functional role of estrogen metabolism in target cells: review and perspectives.
Carcinogenesis (Lond.)
,
19
:
1
1998
.
13
McDonnell D. P., Nawaz Z., O’Malley B. W. In situ distinction between steroid receptor binding and transactivation at a target gene.
Mol. Cell Biol.
,
11
:
4350
1991
.
14
Hyder S. M., Stancel G. M., Chiappetta C., Murthy L., Boettger-Tong H. L., Makela S. Uterine expression of vascular endothelial growth factor is increased by estradiol and tamoxifen.
Cancer Res.
,
56
:
3954
1996
.
15
Nyholm H. C. J., Nielsen A. L., Lyndrup J., Greene G. L., Thorpe S. M. Estrogen receptor determination in endometrial carcinoma: ligand binding assay versus enzyme immunoassay.
Anticancer Res.
,
15
:
649
-654,  
1995
.
16
Pertschuk L. P., Masood S., Simone J., Feldman J. G., Fruchter R. G., Axiotis C. A., Greene G. L. Estrogen receptor immunocytochemistry in endometrial carcinoma: a prognostic marker for survival.
Gynecol. Oncol.
,
63
:
28
-33,  
1996
.
17
Juliano R. L., Varner J. A. Adhesion molecules in cancer: the role of integrins.
Curr. Opin. Cell Biol.
,
5
:
812
1993
.
18
Senger D. R., Ledbetter S. R., Claffey K. P., Papadopoulos-Sergiou A., Perruzzi C. A., Detmar M. Stimulation of endothelial cell migration by vascular permeability factor/vascular endothelial growth factor through cooperative mechanisms involving the αvβ3 integrin, osteopontin, and thrombin.
Am. J. Pathol.
,
149
:
293
1996
.