George M. Stancel

Regulation of vascular endothelial growth factor expression by estrogens and progestins

Estrogens increase the expression of vascular endothelial growth factor (VEGF) mRNA in the rodent uterus. This regulatory effect is rapid, beginning within 1 hr after hormone treatment, dose dependent, and blocked by the pure antiestrogen ICI 182,780. The induction of the transcript is blocked by inhibitors of RNA but not of protein synthesis, and we have recently identified estrogen response elements in the VEGF gene. Collectively, these findings indicate that estrogens regulate uterine VEGF expression at the transcriptional level via the classical nuclear estrogen receptor pathway. Estrogen induction of VEGF occurs in the stromal layer of the rodent uterus, and estradiol induces expression of VEGF transcript levels in cultured human uterine stromal cells. Progestins also induce VEGF expression in the rodent uterus, although the effect is less marked and slower in onset than estrogenic effects. The effect of progestins is blocked by the antiprogestin mifepristone (RU-486), suggesting that it is also mediated by a classical nuclear receptor pathway. In addition, progestins regulate expression of VEGF mRNA and protein in cultured human T47-D breast cancer cells. The development of uterine leiomyomas is associated with exposure to ovarian sex steroids, abnormal uterine bleeding is commonly seen in patients with leiomyomas, and fibroids require an increased vascular supply for their growth. These observations suggest that VEGF and other angiogenic factors may represent potential targets for the treatment and prevention of uterine fibroids.

Interaction of human estrogen receptors α and β with the same naturally occurring estrogen response elements

Abstract Estrogen receptors are derived from two different gene products referred to as estrogen receptor-α (ER-α) and ER-β. Both receptors bind to the consensus estrogen response element (ERE) present in the vitellogenin gene, but their binding to hormone response elements present in other estrogen responsive genes has not been reported yet. Using in vitro expressed human receptors, we now show that ER-β binds to a panel of six endogenous hormone response elements (vitellogenin, c-fos, c-jun, pS2, cathepsin D, and choline acetyltransferase) already known to bind ER-α and confer estrogen inducibility to reporter constructs. Binding of ER-α and ER-β occurred at similar DNA concentrations for some EREs, but different DNA concentrations were required to form complexes of the two receptors with other elements. These results illustrate for the first time by direct receptor–DNA binding studies that both ER-α and ER-β bind to a number of EREs present in endogenous hormone regulated genes, and further suggest that the two forms of the receptor display different patterns of affinities for naturally occurring hormone response elements.

Pharmacological and endogenous progestins induce vascular endothelial growth factor expression in human breast cancer cells

Tumor expansion is dependent on angiogenesis, which is regulated by peptide growth factors of which vascular endothelial growth factor (VEGF) is one of the most selective and potent. VEGF expression is regulated by steroid hormones in a number of systems, including T47‐D human breast cancer cells in which VEGF protein levels are elevated by progestins. In the present study, we investigated the effect of progestins on VEGF mRNA levels in human breast cancer cells. For these experiments, T47‐D cells were exposed to progestins, RNA was prepared for measurement of VEGF transcript levels by Northern blot analysis and VEGF protein in the cell culture media was measured by enzyme‐linked immunosorbent assay. Basal expression of VEGF mRNA is low in these cells, and is rapidly induced following exposure to progestins, reaching a maximum induction of 2‐ to 5‐fold between 3 and 6 hr after hormone addition. This induction was inhibited by the antiprogestin RU‐486 indicating that it is progesterone receptor (PR) dependent. Transcripts for VEGF165 and VEGF121 were the two major spliced forms of VEGF mRNA that were detected by reverse transcription‐polymerase chain reaction in basal and progestin‐stimulated T47‐D cells. Maximum induction of VEGF mRNA was achieved with 10−8 M progesterone, and induction was hormone specific, as estrogens, glucocorticoids, and androgens were without effect. Actinomycin D completely abolished the induction of VEGF transcript levels by progestins, suggesting that this response involves de novo mRNA synthesis, but puromycin did not inhibit induction, suggesting that this effect does not require protein synthesis. This report demonstrates that progestins stimulate VEGF mRNA levels and raises the possibility that anti‐progestins may be useful to inhibit proliferation and metastasis in some human breast cancers by blocking VEGF production.

Regulation of uterine epidermal growth factor (EGF) receptors by estrogen in the mature rat and during the estrous cycle

Previous work has shown that the immature rat uterus contains epidermal growth factor (EGF) receptors and that tissue levels of this receptor are increased by the administration of exogenous estrogens. This study was undertaken to determine if estrogen administration also elevated EGF receptor levels in the mature animal and if the growth factor receptor levels varied in concert with endogenous estrogens throughout the estrous cycle. In the mature, castrate rat administration of estradiol, but not non-estrogenic steroids, causes a 2-3-fold elevation of uterine EGF receptors as judged by ligand binding. This increase is maximum in 18 h and is due to an increase in the number of binding sites. In cycling animals EGF receptor levels are low at metestrus, rise at diestrus, reach a maximum (approximately twice metestrus values) at proestrus, and then return at estrus to metestrus levels. These changes in EGF receptor levels parallel changes in plasma estrogens and occupied nuclear estrogen receptor reported by other workers. These results indicate that uterine EGF receptors are increased by exogenous estrogens in both mature and immature animals, and support a physiological role for estrogens in the regulation of this growth factor receptor.

DDT supports the growth of an estrogen-responsive tumor. Mammary tumor growth; pesticides

MT2 cells, a clonal cell line of MTW9/PL cells originally obtained from a mammary adenocarcinoma, from estrogen-responsive tumors in Wistar-Furth rats. o,p-DDT supports MT2 tumor growth at a rate similar to 17 beta-estradiol. This effect of o,p-DDT is dose-dependent and specific, since the DDT congener p,p-DDD, which does not bind to tumor estrogen receptors, does not support tumor growth. This is the first demonstration that DDT can support the growth of an estrogen-responsive tumor.

Estrogen regulates expression of the jun family of protooncogenes in the uterus

Treatment of immature female rats with estradiol increases uterine levels of c-jun and jun-B mRNAs approx. 10-fold. This effect is specific for estrogenic steroids. The induction of jun transcripts is blocked by actinomycin D but not puromycin, suggesting that the hormonal effect is due at least in part to transcriptional activation. The hormone effect is rapid and peak levels of jun mRNAs are seen within 3 h after treatment. Inductions of jun and fos transcripts in the uterus by estradiol exhibit similar dose response curves (maximum responses at 4 micrograms/kg). Estradiol also elevates uterine levels of jun-D, and this induction is insensitive to puromycin. In vivo treatment with the phorbol ester TPA rapidly elevates uterine levels of fos, jun, and myc transcripts, indicating that expression of these protooncogenes is under non-estrogenic as well as estrogenic regulation in this target tissue. These results suggest that multiple members of the jun and fos protooncogene families may play a role in amplifying the uterine response to estrogens.

Hormonal carcinogenesis: Separation of estrogenicity from carcinogenicity*

Estrogens are known to induce tumors in several animal species. To understand the mechanism of hormonal carcinogenesis, estrogen-induced renal carcinoma in male Syrian hamsters was investigated using estradiol and 2-fluoroestradiol. The biological activities of the latter steroid were compared with those of the natural hormone, because of the reduced metabolic conversion of 2-fluoroestradiol to catechol estrogen metabolites. 2-Fluoroestradiol was administered to male Syrian hamsters at three times the dose (60 mg) of estradiol (20 mg, positive control) by s.c. implantation. After 7 months, 75% of the estradiol-treated hamsters had kidney tumors, while in animals exposed to 2-fluoroestradiol renal carcinoma could not be detected. The reduced tumor incidence by the fluorinated steroid is not due to a lack of estrogenic potency. In the test animals, pituitary LH concentrations matched those measured in estradiol-treated hamsters and the reduction in testes weights was comparable. Furthermore, in immature female rats, uterine wet weight increases illustrate that 2-fluoroestradiol is a potent estrogen. The observed increases in uterine weight were shown to be accompanied by increases in protein and DNA synthesis comparable to those observed in estradiol-treated animals. 2-Fluoroestradiol stimulated growth of H-301 cells in vivo. These cells are estrogen-dependent for growth and are derived from the primary hamster kidney tumor. The results indicate that hormonal activity and carcinogenicity of estrogens are separable properties.

Estrogenic activity of DDT: Estrogen‐receptor profiles and the responses of individual uterine cell types following o,p'‐DDT administration

The administration of o,p-dichlorodiphenyltrichloroethane (o,p-DDT) to immature rats stimulates DNA synthesis and cell division in the uterine luminal epithelium (LE), stroma (S), and myometrium (M). The time course of DNA synthesis/cell division in the S and M is similar following administration of o,p-DDT or 17 beta-estradiol, but the maximum response following pesticide treatment is only 70% of that produced by the hormone. In the LE both compounds yield the same maximum response, but the time course of DNA synthesis/cell division is delayed following o,p-DDT administration relative to 17 beta-estradiol treatment. The patterns of estrogen receptor retention in uterine nuclei following o,p-DDT administration are prolonged relative to those observed after 17 beta-estradiol treatment. o,p-DDT thus produces the uterine hyperplasia characteristic of estrogens, but the magnitude and timing of the response is dependent on the specific cell type observed and is different from that produced by estradiol.

Separation of multiple forms of cyclic nucleotide phosphodiesterases from rat brain by isoelectrofocusing

Abstract Multiple forms of cyclic nucleotide phosphodiesterase(s) (EC 3.1.4-) have been investigated using isoelectric focusing techniques. Six distinct peaks of cyclic AMP phosphodiesterase activity are apparent in a 105 000 × g soluble supernatant fraction of sonicated rat cerebellum. These peaks, designated A–F, have isoelectric points (pI values) of 4.4, 4.8, 5.0, 6.1, 8.3 and 9.0, respectively, varying affinities for cyclic AMP and cyclic GMP, different kinetic behavior, and divergent subcellular localization. Peaks B, C and D contain appreciable cyclic GMP phosphodiesterase activities, while Peaks A, E and F hydrolyze little or no cyclic GMP. Kinetic analysis of five of the focused peaks showed non-linear Lineweaver-Burk plots closely approximating those of the original cerebellar homogenate. Discontinuous sucrose gradient fractionation before isoelectric focusing indicates that Peaks B, D and E are cytosolic forms and the others appear particulate in nature. In contrast to multiple forms separated by preparative polyacrylamide gel electrophoresis, activity peaks separated by isoelectric focusing do not respond to an endogenous activator of the cyclic AMP phosphodiesterase activity. Since the isoelectric focusing technique has preparative and analytical advantages over other electrophoretic methods, it provides a good procedure to investigate components of the complex cyclic nucleotide phosphodiesterase enzyme system.

Presence of an estradiol response region in the mouse c-fos oncogene

We have previously shown that the intracellular content of c-fos mRNA is rapidly induced (within 1 to 3 hours) in ovariectomized rat or mouse uteri following administration of estradiol. This induction is sensitive to actinomycin D but not to protein synthesis inhibitor puromycin, indicating an effect of estradiol at the transcriptional level, possibly mediated by the estrogen receptor. We have used transient transfection assays with defined regions of the mouse c-fos gene ligated to a reporter plasmid expressing chloramphenicol acetyl transferase to study regulation of this gene by estrogens. These recombinants were transfected in two different estrogen-responsive cell lines, GH4 and MCF-7, and stimulated with estradiol. A two- to five-fold induction of chloramphenicol acetyl transferase activity was observed with a construct containing the intact c-fos promoter and 351 bases of 5-flanking sequence (-351/+44). A similar induction by estrogen is observed with the endogenous c-fos gene in the two cell lines as determined by RNA blot analysis. Estrogen induction is lost when a construct containing -135/+44 region of the c-fos gene is transfected. Plasmid containing the consensus estrogen response element GGTCAnnnTGACC derived from vitellogenin gene is induced 10- to 50-fold in both estrogen-responsive cell lines. Under identical conditions, the oligonucleotide containing the perfect palindrome GGTCTnnnAGACC, present around the -209 region of the c-fos gene, is completely silent when transfected under the control of thymidine kinase promoter. Additional transfection analysis with a number of c-fos promoter constructs has narrowed the estrogen response region to within the -278 to -135 region upstream of the c-fos promoter.