- Email: [email protected]
ICI 182,780 acts as a partial agonist and antagonist of estradiol effects in specific cells of the sheep uterus
Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281– 287 www.elsevier.com/locate/jsbmb
ICI 182,780 acts as a partial agonist and antagonist of estradiol effects in specific cells of the sheep uterus Jane A. Robertson a, Yuhua Zhang a, Nancy H. Ing a,b,* b
a Department of Animal Science, Texas A&M Uni6ersity, College Station, TX 77843 -2471, USA Veterinary Anatomy and Public Health and Center for Animal Biotechnology, Institute of Biosciences and Technology, Texas A&M Uni6ersity, College Station, TX 77843 -2471, USA
Received 6 March 2000; accepted 28 February 2001
Abstract We assessed the ability of ICI 182,780 (ICI) to block the estradiol (E2) responses of genes within the sheep uterus. Ovariectomized ewes in the ‘ICI +E2’ treatment group received a uterine infusion with 10 − 7 M ICI for 14 h, an injection of 50 mg E2 6 h after the infusion started, and were hysterectomized 18 h postinjection. Other groups received only ICI or E2, or neither treatment (‘Con’). Both E2 and ICI increased the wet weight of dissected endometrium: averaging 10.0 91.2 g for ICI +E2, ICI, and E2 groups compared to 6.8 90.6 g for Con. Slot blot analyses of endometrial RNA showed that estrogen receptor-a (ER), progesterone receptor (PR), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), cyclophilin, actin and c-fos mRNAs responded to E2 treatment: the first five increased an average of 60% while the last decreased 38%. In situ hybridization identified more subtle ICI effects: agonistic up-regulation of GAPDH mRNA in superficial endometrial cells, and antagonistic down-regulation of ER and PR mRNAs in the inner layer of the myometrium. Thus, we conclude that the agonist versus antagonist effects of ICI relative to those of E2 are a function of the gene examined as well as the specific cell within the uterus. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: ICI 182,780; Estradiol; Uterus
1. Introduction Powerful antiestrogen drugs, such as ICI 182,780 (‘ICI,’ also known as Faslodex) have been developed for their ability to inhibit estrogen-driven growth of breast cancer cell lines [1]. In the uterus, however, results of some antiestrogen treatments mimic estrogen effects. Tamoxifen, the antiestrogen most widely used in human medicine, stimulates uterine growth in rodents, pigs, and women [2,3] and is reported to increase the incidence of endometrial cancer [1]. ICI is a relatively new ‘Type II’ antiestrogen [4]. Although ICI binds uterine ER protein with affinity similar to that of
E2 [5], it has little or no uterotrophic activity in rodents, pigs and women, and appears to antagonize E2 induction of endometrial growth [6–10]. We have identified several genes, including ER, PR, GAPDH, cyclophilin, and c-fos, that are up-regulated in sheep endometrium by a single physiological dose of E2 [11,12]. Previously, we showed that tamoxifen also up-regulates expression of these endometrial genes [11]. Here, we examined whether ICI acts as an agonist or antagonist of E2 effects on growth and gene expression in the uteri of ovariectomized sheep.
2. Materials and methods Abbre6iations: Con, control; E2, estradiol; ER, estrogen receptor-a; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; ICI, ICI 182,780; PR, progesterone receptor. * Corresponding author. Tel.: +1-979-8622790; fax: + 1-9798623399. E-mail address: [email protected] (N.H. Ing).
2.1. Animal treatments and sample collection Crossbred Rambouillet ewes (O6is aries) were ovariectomized during the breeding season after exhibiting breeding cycles of normal duration (16 –18
0960-0760/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0960-0760(01)00061-9
282
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
days). Four weeks after ovariectomy, ewes were randomly assigned to four treatment groups (n = 4 ewes per group). Time of E2 injection (50 mg in charcoalstripped corn oil, intramuscular) was 18 h prior to hysterectomy, a time of rapid ER mRNA accumulation in endometrium [13]. ICI was administered for 14 h starting 24 h before hysterectomy (6 h prior to E2 treatment) and ending 10 h prior to hysterectomy (8 h subsequent to the E2 challenge) in order to competitively occupy ER protein. ICI (10 − 7 M ICI in 0.1% ovine serum albumin in phosphate-buffered saline) was delivered by infusing uteri via catheters seated in the tips of both uterine horns with either at a rate of 3 ml/h per horn. This rate is based on volumes known to be absorbed over time without causing uterine distension. At hysterectomy, uterine cross-sections were taken from the middle of one uterine horn just distal to the external bifurcation and fixed in 4% (w/w) paraformaldehyde before being embedded in Ameraffin (Baxter Diagnostics; Deerfield, IL). Endometrium was dissected from the remaining uterus, minced, snap-frozen in liquid nitrogen and stored at −80°C for RNA slot blot analyses. All animal procedures were approved by the Texas A&M University Laboratory Animal Care and Use Committee. Unless otherwise indicated, chemicals were purchased from Sigma (St Louis, MO).
2.2. Synthesis of cRNA probes For probing endometrial RNA immobilized on slot blots, cRNA probes were generated from ovine ER, PR, GAPDH, and c-fos cDNA as described previously [11,13]. The cRNA probes for cyclophilin and actin were generated from pTRI-cyclophilin-Human and pTRI-b-actin-Human (Ambion; Austin, TX). Briefly, antisense cRNA probes were generated by in vitro transcription with T7 RNA polymerase (Maxiscript kit; Ambion) and [32P]UTP (New England Nuclear; Boston, MA). Unincorporated radionucleotides were removed on spin columns (Bio 101; La Jolla, CA). For in situ hybridization, antisense and sense ER, PR, and GAPDH cRNA probes were produced as above except that [35S]UTP was substituted for [32P]UTP and the sense probes were synthesized with SP6 RNA polymerase.
2.3. Preparation of RNA samples and RNA slot blot analyses Total cellular RNA was purified from quadruplicate tissue samples (0.1 g each) with the Tripure Reagent (Roche Molecular Biochemicals; Indianapolis, IN). Absorbances at 260 nm were measured in duplicate and 5 mg total cellular RNA were loaded
on replicate slot blots with a BioRad vacuum manifold as previously described [11]. High stringency was maintained throughout blot hybridization (55°C) and washing {68°C in 0.1× SSC (1 × SSC = 150 mM NaCl + 15 mM Na citrate) and 0.1% sodium dodecyl sulfate followed by treatment with 1 mg/ml RNase A (Roche Molecular Biochemicals) for 10 min at room temperature}. Hybridization signals on blots were counted on an InstantImager (Packard; Meriden, CT). Prior Northern analyses using this protocol and these probes detected only bands of the expected sizes for each of the mRNAs analyzed [12].
2.4. In situ hybridization Serial 7-mm cross-sections from the uterus of each ewe were placed on Superfrost Plus slides (Curtin Matheson Scientific; Houston, TX). Within 1 week of sectioning, cross-sections were hybridized with one of six radiolabeled probes: sense and antisense ER, PR, and GAPDH cRNAs as described previously [14]. Slides were developed after 6 weeks of exposure to autoradiographic emulsion (NTB-2 from Kodak; Rochester, NY). Sections were counterstained lightly with hematoxylin. Densities of silver grains over different uterine cell types were quantitated on a Reichert Microstar IV microscope and NIH Image software version 1.61. Pixel densities were averaged on each uterine cross-section for each antisense probe over the following tissue compartments: luminal epithelium (LE), stratum compactum (SC, dense stroma just beneath the LE), superficial glandular epithelium (SGE), the adjacent superficial stroma (SS), deep glandular epithelium (DGE), deep stroma (DS), inner circular myometrium (Mi) and outer longitudinal myometrium (Mo) [14]. For each compartment, pixel densities on sections probed with sense cRNAs (negative controls for non-specific hybridization) were subtracted as background signals.
2.5. Statistical analyses Quantitative data were analyzed by least squares ANOVA using the General Linear Model procedures of SAS [15]. Results from treatment groups were compared: E2 versus Con for E2 effects, ICI versus Con for agonist effects of ICI, and ICI+ E2 versus E2 for antagonistic effects of ICI. Data are presented as least squares means and S.E. for treatment groups. Messenger RNA ‘concentrations’ from slot blots are expressed as a percentage of the average hybridization signal of the Con group. In situ hybridization signals are reported as pixel densities. Level of statistical significance is PB0.05 unless otherwise stated.
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
283
to the small, pale tracts of the Con ewes and the large, turgid, reddened tracts of the E2 ewes. Uteri from the ICI+ E2 treatment group were indistinguishable in outward appearance from those treated with E2 alone. Wet weights of dissected endometrium ranged from 5.7 to 12.7 g. The wet weights of endometrium were similar in ewes from E2, ICI and ICI + E2 groups and were an average of 50% greater compared to those of control ewes (Fig. 1). These results are consistent with ICI acting as an E2 agonist, not as an antagonist.
3.2. E2 regulates expression of endometrial genes
Fig. 1. Both E2 and ICI treatments increased dissected endometrial weights. Adult ewes were treated with E2, ICI or ICI +E2 as described in the text (n =4 ewes per treatment group). Wet weights of endometrium are reported in g as least squares means and S.E. for treatment groups. E2 and/or ICI treatments increased endometrial weights compared to controls (‘Con’) ewes (asterisks indicate differences of P =0.06).
3. Results
3.1. ICI affects gross uterine characteristics similar to E2 Whole uteri of ewes treated with ICI alone were intermediate in gross outward appearance compared
Gross endometrial levels of ER, PR, GAPDH, cyclophilin, actin and c-fos mRNAs were measured on slot blots. Quantitation of the hybridization signals indicated that E2 treatment up-regulated the expression of ER mRNA 70% relative to Con values (Fig. 2). E2 also up-regulated PR, GAPDH, cyclophilin and actin mRNA levels in endometrium with effects ranging from 37% (PR, PB 0.06) to 85% (actin). E2 had the opposite effect on c-fos mRNA: it down-regulated it to 38% of Con values. ICI, alone or in combination with E2, did not affect the concentrations of the mRNAs measured (P\ 0.4). However, antagonism was suggested by the fact that, for all five E2 up-regulated mRNAs, the hybridization signals in samples from ICI+ E2 ewes appeared to be an average of 12% lower than those from E2 treatment group ewes.
Fig. 2. Estradiol treatment affected mRNA levels in endometrium. Total cellular RNA isolated from quadruplicate samples of endometrium analyzed on slot blots for ER, PR, GAPDH, cyclophilin (‘Cyclophil’), actin and c-fos mRNAs. Results are presented as least squares means and S.E., expressed as percentage of the mean value of Con. E2 (filled bars) increased all mRNA concentrations, except it decreased that of c-fos (P5 0.06). ICI alone (hatched bars) or with E2 (stippled bars) had no significant effect, agonistic or antagonistic to that of E2.
284
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
Fig. 3. In situ hybridization of GAPDH mRNA indicates gene expression in specific uterine cell compartments. Representative views of in situ hybridization results on uterine cross-sections from ewes in Con, E2, ICI and ICI +E2 treatment groups are shown in panels A – D, respectively. Black silver grains on the brightfield images indicate in situ hybridization of the antisense GAPDH cRNA probe. Nuclei, lightly stained with hematoxylin, are visible primarily in panel A due to its lower hybridization signals. Panels span the upper (superficial) half of endometrium from luminal epithelium (LE) to superficial endometrial glands (SGE) extending down from the uterine lumen, at far right. The bar represents 100 mm.
3.3. In situ hybridization analyses show cell-specific effects of ICI To examine effects at the cellular level, ER, PR and GAPDH cRNA probes were used for in situ hybridization. The up-regulation of ER and PR mRNAs have been extensively analyzed previously [14]. However, this is the first in situ characterization of GAPDH mRNA in sheep uteri (Fig. 3). It demonstrated that GAPDH mRNA was more strongly expressed in epithelial cells of the endometrium compared to endometrial stroma and myometrium. The low expression levels of GAPDH mRNA in Con ewes (Fig. 3A) were strongly up-regulated by E2 in the stromal cells (SC and SS) below the epithelial cells lining the uterine lumen (LE), as well as within the epithelium of the upper glands (SGE; Fig. 3B). ICI acted as an E2 agonist in up-regulating GAPDH mRNA in SC and SGE cell types (Fig. 3C) and did not inhibit the E2 up-regulation described above (Fig. 3D). Hybridization signals for ER, PR and GAPDH mRNAs were quantitated over cell compartments spanning the endometrium and myometrium. The strengths of the hybridization signals were, in general, strongest for ER mRNA and weakest for PR mRNA (Fig. 4). For all three mRNAs analyzed, expression was lowest in the
less dense stromal compartments (SS and DS). The only significant effect of E2 was to increase ER mRNA 207% in Mi (Fig. 4, top panel). However, all compartments showed a trend in increasing ER mRNA in response to E2. ICI treatment, when used alone, had no effect. When used in conjunction with E2, however, ICI inhibited the rise in ER mRNA in Mi. Results of PR mRNA in situ hybridization were similar to those for ER mRNA: E2 up-regulated PR mRNA 221% in the Mi cell compartment and ICI inhibited the up-regulation (Fig. 4, middle panel). Quantitation of the in situ hybridization of GAPDH mRNA demonstrated that E2 increased GAPDH mRNA in SC, SGE, SS, DGE and Mi compartments between 142 and 442% (Fig. 4, bottom panel). ICI treatment mimicked that of E2 in up-regulating GAPDH mRNA in SC and SGE cells 208 and 142%, respectively. ICI did not antagonize E2 effects on GAPDH gene expression.
4. Discussion The uterus, and especially the endometrium, is exquisitely responsive to E2. We showed previously [11] that a single 50-mg dose of E2 increased uterine wet
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
weight, a classic bioassay for estrogenic effects. In this study, ICI acted like E2 by increasing uterine redness and endometrial weight. ICI also acted as an estrogen agonist on uterine parameters involved in luteolysis in the only other published study employing adult sheep [16]. Uterine gland growth, responsive to E2 treatment
285
in neonatal lambs, pigs and calves ([8], and references therein), also increased in neonatal lambs with ICI treatment (125 mg/kg per day for 14 days; Dr Thomas E. Spencer, pers. commun.). The same ICI treatment of neonatal pigs had no effect on uterine gland growth [8]. The fact that three studies with sheep have found
Fig. 4. Quantitation of in situ hybridization results for ER, PR and GAPDH mRNAs indicates changes in gene expression in specific uterine cell compartments. Pixel densities measured by light microscopy and NIH Image software are reported for ER, PR and GAPDH (top, middle, and bottom panels, respectively). The different uterine cell compartments analyzed are: luminal epithelium (LE), stratum compactum (SC, dense stroma just beneath the LE), superficial glandular epithelium (SGE), the adjacent superficial stroma (SS), deep glandular epithelium (DGE), deep stroma (DS), inner circular myometrium (Mi) and outer longitudinal myometrium (Mo). Least squares means and S.E. of pixel densities on uterine cross-sections from Con, E2, ICI and ICI + E2 treatment groups are reported in open, filled, hatched and stippled bars, respectively. Asterisks (*) and number symbols (c ) indicate differences compared to Con and E2 groups, respectively.
286
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
uterotropic actions of ICI may indicate a true species difference of sheep with the rat, pig, monkey, and human species, in which ICI acts as a pure estrogen antagonist [4,6–10,13]. The amino acid sequences of sheep, pig and human ER proteins (GenBank accession numbers Z49257, Z37167 and X03635, respectively) are highly conserved: 95% (sheep/pig) and 91% (sheep/human) identical overall. However, the C-terminus is much less conserved: 72% (sheep/pig) and 63% (sheep/ human) identical over the last 43 residues. Since the C-terminus of ER affects ligand discrimination [17], perhaps these differences alter the effect of ICI with the sheep ER compared to ICI with the pig and human ER proteins. We examined ICI effects, with and without E2, on a battery of genes known to respond to E2 in the sheep endometrium [11,12]. Consistent with previous results, E2 increased gross endometrial expression of ER, PR, GAPDH, and cyclophilin genes. Additionally, we report for the first time that E2 up-regulates actin mRNA in sheep endometrium. Although many regard GAPDH, cyclophilin and actin genes as unresponsive, ‘house-keeping’ genes, others have reported their upregulation by E2 in the rodent uterus [18,19]. Thus, care should be taken in normalizing data to expression of a ‘house-keeping’ gene, especially during a vigorous response like that of the uterus under the influence of E2. In this study, E2 down-regulated c-fos mRNA. Our other studies in ovariectomized ewes have shown early c-fos up-regulation [11] and later down-regulation [12]. This biphasic response of the c-fos gene has been explained by early induction of transcription by activated ER protein followed by autologous down-regulation by the c-fos protein itself [20]. Tamoxifen, a predecessor of ICI, acted like E2 in up-regulating ER and GAPDH gene products in sheep endometrium [11]. In this study, ICI did not act as an agonist or antagonist on gross levels of any of the E2-regulated gene products. However, in situ hybridization allowed dissection of the separate uterine cell compartments for individual assessment of their changes in gene expression in response to E2 and ICI. With it, we are the first to report that the GAPDH gene is expressed predominantly in the endometrial epithelium and myometrial cells of the sheep uterus. E2 up-regulated GAPDH mRNA levels in most uterine cell types. In addition, ICI acted like E2 in up-regulating GAPDH mRNA in the superficial stroma and glandular epithelium of the endometrium. Other examples of ICI acting as an estrogen agonist have been described on a subset of estrogen-responsive gene promoters in uterine cell lines and primary cultures [21– 23]. There is also evidence that ICI can activate these promoters with estrogen receptor-b. However, estrogen receptor-b gene expression in the uterus is very low relative to that of ER-a [24,25].
Regulation of ER and PR genes by E2 in specific uterine cells was characterized previously [14]. This report is consistent with that study: E2 markedly enhanced ER and PR mRNA levels in the innermost layer of myometrium. In this robustly responsive uterine compartment, the in situ hybridization data revealed that ICI antagonized the E2 up-regulation of both ER and PR mRNAs. These data agree with the majority of studies employing ICI as an estrogen antagonist in vivo. For example, ICI antagonized the estrogen induction of PR, c-fos, lactoferrin and vascular endothelial growth factor genes in mouse uterus [26,27]. Those studies used ratios of ICI to E2 ranging from 25 to 100, which brackets our estimated ratio of 70 (based on the assumption that the endometrium equilibrates to 10 − 7 M ICI in the infusate and retains 850 pg E2 per g ($ ml) endometrium from the injection [11]). The fact that the E2 antagonism occurred in myometrium, the furthest tissue from the luminal delivery site of the ICI, confirms that the inhibitor delivered in the 42 ml of total infusate penetrated the entire uterus (25–40 g wet wt. [11]). One explanation of ICI antagonism of E2 effects is that ICI down-regulates levels of ER protein [1]. To see if ER protein levels were preferentially changed in the cell compartments that responded to ICI, we performed immunohistochemistry with the anti-ER H222 antibody on uterine cross-sections. Although, we saw no differences in endometrial or myometrial ER protein levels with ICI treatment (data not shown), it is possible that prior down-regulation of ER protein levels recovered by the time of hysterectomy. However, some reported increases in uterine ER protein during negative effects on uterine growth in mice treated with ICI [28]. It is important to note that estrogens rapidly down-regulate levels of immunoreactive ER protein in cells [13,29] so they may not directly correlate temporally to the action of E2 in a cell, especially during acute treatments. Rather, specific cells may activate cofactors, corepressors, and/or convergent signaling pathways differentially in response to E2 and/or ICI [21,30]. We conclude that ICI effects on estrogen-regulated gene expression, a balance between agonism and antagonism, are strongly dependent upon the species, gene, and cell type being studied [1,11,30].
Acknowledgements The authors gratefully acknowledge the generous gift of ICI 182,780 from Dr Alan Wakeling of Zeneca Pharmaceuticals, without which the study would not have been possible. Also instrumental were the technical assistance of Cindy Balog and the Image Analysis core facility of Dr Robert C. Burghardt and the Center for Environmental and Rural Health. Research was
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
conducted by the Texas Agricultural Experiment Station, with funding provided by award c 95-37203-2182 to NHI from the USDA NRI-CGP.
References [1] J.I. MacGregor, V.C. Jordan, Basic guide to mechanisms of antiestrogen action, Pharm. Rev. 50 (1998) 151 – 196. [2] P.G. Satyaswaroop, Estrogenic and antiestrogenic actions of tamoxifen in the female reproductive tract, in: R. Lindsay, D.W. Dempster, V.C. Jordan (Eds.), Estrogens and Antiestrogens, Lippincott-Raven, Philadelphia, PA, 1997, pp. 2121 –2221. [3] C.L. Lin, H.L. Buttle, Effect of oestradiol benzoate and tamoxifen on the growth and induction of progesterone receptors in the uterus and mammary gland of immature pigs, J. Endocrinol. 130 (1991) 259 –265. [4] A.E. Wakeling, M. Dukes, J. Bowler, A potent specific pure antiestrogen with clinical potential, Cancer Res. 51 (1991) 3867 – 3873. [5] C. Martel, L. Provencher, X. Li, A. St. Pierre, G. Leblanc, S. Gauthier, Y. Merand, F. Labrie, Binding characteristics of novel nonsteroidal antiestrogens to the rat uterine estrogen receptors, J. Steroid Biochem. Mol. Biol. 64 (1998) 199 –205. [6] W.S. Branham, R. Fishman, R.D. Streck, K.L. Medlock, J.J. DeGeorge, D.M. Sheehan, ICI 182,780 inhibits endogenous estrogen-dependent rat uterine growth and tamoxifen-induced developmental toxicity, Biol. Reprod. 54 (1996) 160 –167. [7] L. Sahlin, H. Eriksson, Influence of blockers for the estrogen receptor (ER) and type 1 IGF-receptor on the levels of ER mRNA and IGF-1 mRNA in the rat uterus, J. Steroid Biochem. Mol. Biol. 58 (1996) 359 –365. [8] B.J. Tarleton, A.A. Wiley, F.F. Bartol, Endometrial development and adenogenesis in the neonatal pig: effects of estradiol valerate and the antiestrogen ICI 182,780, Biol. Reprod. 61 (1999) 253 –263. [9] M. Dukes, J.C. Waterton, A.E. Wakeling, Antiuterotrophic effects of the pure antioestrogen ICI 182,780 in adult female monkeys (Macaca nemestrina): quantitative magnetic resonance imaging, J. Endocrinol. 138 (1993) 203 – 209. [10] E.J. Thomas, P.L. Walton, N.M. Thomas, M. Dowsett, The effects of ICI 182,780, a pure anti-oestrogen, on the hypothalamic– pituitary –gonadal axis and on endometrial proliferation in pre-menopausal women, Hum. Reprod. 9 (1996) 1991 –1996. [11] J.A. Robertson, S. Bhattacharyya, N.H. Ing, Tamoxifen up-regulates oestrogen receptor-alpha, c-fos and glyceraldehyde 3phosphate-dehydrogenase mRNAs in ovine endometrium, J. Steroid Biochem. Mol. Biol. 67 (1999) 285 –292. [12] K. Zou, N.H. Ing, Oestradiol up-regulates oestrogen receptor, cyclophilin, and glyceraldehyde phosphate dehydrogenase mRNA concentrations in endometrium but down-regulates them in liver, J. Steroid Biochem. Mol. Biol. 64 (1998) 231 –237. [13] N.H. Ing, T.E. Spencer, F.W. Bazer, Estrogen enhances endometrial estrogen receptor gene expression by a posttranscriptional mechanism in the ovariectomized ewe, Biol. Reprod. 54 (1996) 591 – 599.
287
[14] N.H. Ing, M.B. Tornesi, Estradiol up-regulates estrogen receptor and progesterone receptor gene expression in specific ovine uterine cells, Biol. Reprod. 56 (1997) 1205 – 1215. [15] SAS User’s Guide: Basics. SAS Institute, Cary, 1982. [16] H. Al-Matubsi, R.J. Fairclough, G. Jenkin, Oestrogenic effects of ICI 182,780, a putative anti-oestrogen on the secretion of oxytocin and prostaglandin F2a during oestrous cycle in the intact ewe, Anim. Reprod. Sci. 51 (1998) 81 – 96. [17] B.S. Katzenellenbogen, M.M. Montano, P. Le Goff, D.J. Schodin, W.L. Kraus, B. Bhardwaj, N. Fujimoto, Antiestrogens: mechanisms and actions in target cells, J. Steroid Biochem. Mol. Biol. 53 (1995) 387 – 393. [18] M. Rosser, L. Chorich, E. Howard, P. Zamorano, V.B. Mahesh, Changes in rat uterine estrogen receptor messenger ribonucleic acid levels during estrogen- and progesterone-induced estrogen receptor depletion and subsequent replenishment, Biol. Reprod. 48 (1993) 89 – 98. [19] L. Sahlin, Dexamethasone attenuates the oestradiol-induced increase in IGF-1 mRNA in the rat uterus, J. Steroid Biochem. Mol. Biol. 55 (1995) 9 – 15. [20] M. Schuchard, J.P. Landers, N.P. Sandhu, T.C. Spelsberg, Steroid hormone regulation of nuclear proto-oncogenes, Endocr. Rev. 14 (1993) 659 – 669. [21] E. Castro-Rivera, S. Safe, Estrogen and antiestrogen-responsiveness of HEC1A endometrial adenocarcinoma cell in culture, J. Steroid Biochem. Mol. Biol. 64 (1998) 287 – 295. [22] P.S. Jones, E. Parrott, I.N.H. White, Activation of transcription by estrogen receptor a and b is cell-type and promoter-dependent, J. Steroid Biochem. Mol. Biol. 274 (1999) 32008 –32014. [23] K. Paech, P. Webb, G.G.J.M. Kuiper, S. Nilsson, J.-A. Gustafsson, P.J. Kushner, T.S. Scanlan, Differential ligand activation of estrogen receptors ERa and ERb at AP1 sites, Science 277 (1997) 1508 – 1510. [24] G.G.J.M. Kuiper, B. Carlsson, K. Grandien, E. Enmark, J. Haggblad, S. Nilsson, J.-A. Gustafsson, Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptor a and b, Endocrinology 138 (1997) 863 –870. [25] H. Wong, B. Masironi, H. Eriksson, L. Sahlin, A comparative study of estrogen receptors a and b in the rat uterus, Biol. Reprod. 62 (2000) 310 – 317. [26] S.K. Das, J. Tan, D.C. Johnson, S.K. Dey, Differential spatiotemporal regulation of lactoferrin and progesterone receptor gene in the mouse uterus by primary estrogen, catechol estrogen, and xenoestrogen, Endocrinology 139 (1998) 2905 – 2915. [27] S.M. Hyder, C. Chiapetta, L. Murthy, G.M. Stancel, Selective inhibition of estrogen-regulated gene expression in vivo by the pure antiestrogen ICI 182,780, Cancer Res. 57 (1997) 2547 – 2549. [28] C. Martel, C. Labrie, A. Belanger, S. Gauthier, Y. Merand, X. Li, L. Provencher, B. Candas, F. Labrie, Comparison of the effects of the new orally active antiestrogen EM-800 with ICI 182 780 and toremifene on estrogen-sensitive parameters in the ovariectomized mouse, Endocrinology 139 (1998) 2486 –2492. [29] Z. Nawaz, D.M. Lonard, A.P. Dennis, C.L. Smith, B.W. O’Malley, Proteasome-dependent degradation of the human estrogen receptor, Proc. Natl. Acad. Sci. USA 96 (1999) 1858 – 1862. [30] B.S. Katzenellenbogen, M.M. Montano, K. Ekana, M.E. Herman, E.M. McInerney, Antiestrogens: mechanisms of action and resistance in breast cancer, Breast Cancer Res. Treat. 44 (1997) 23 – 38.
ICI 182,780 acts as a partial agonist and antagonist of estradiol effects in specific cells of the sheep uterus Jane A. Robertson a, Yuhua Zhang a, Nancy H. Ing a,b,* b
a Department of Animal Science, Texas A&M Uni6ersity, College Station, TX 77843 -2471, USA Veterinary Anatomy and Public Health and Center for Animal Biotechnology, Institute of Biosciences and Technology, Texas A&M Uni6ersity, College Station, TX 77843 -2471, USA
Received 6 March 2000; accepted 28 February 2001
Abstract We assessed the ability of ICI 182,780 (ICI) to block the estradiol (E2) responses of genes within the sheep uterus. Ovariectomized ewes in the ‘ICI +E2’ treatment group received a uterine infusion with 10 − 7 M ICI for 14 h, an injection of 50 mg E2 6 h after the infusion started, and were hysterectomized 18 h postinjection. Other groups received only ICI or E2, or neither treatment (‘Con’). Both E2 and ICI increased the wet weight of dissected endometrium: averaging 10.0 91.2 g for ICI +E2, ICI, and E2 groups compared to 6.8 90.6 g for Con. Slot blot analyses of endometrial RNA showed that estrogen receptor-a (ER), progesterone receptor (PR), glyceraldehyde 3-phosphate dehydrogenase (GAPDH), cyclophilin, actin and c-fos mRNAs responded to E2 treatment: the first five increased an average of 60% while the last decreased 38%. In situ hybridization identified more subtle ICI effects: agonistic up-regulation of GAPDH mRNA in superficial endometrial cells, and antagonistic down-regulation of ER and PR mRNAs in the inner layer of the myometrium. Thus, we conclude that the agonist versus antagonist effects of ICI relative to those of E2 are a function of the gene examined as well as the specific cell within the uterus. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: ICI 182,780; Estradiol; Uterus
1. Introduction Powerful antiestrogen drugs, such as ICI 182,780 (‘ICI,’ also known as Faslodex) have been developed for their ability to inhibit estrogen-driven growth of breast cancer cell lines [1]. In the uterus, however, results of some antiestrogen treatments mimic estrogen effects. Tamoxifen, the antiestrogen most widely used in human medicine, stimulates uterine growth in rodents, pigs, and women [2,3] and is reported to increase the incidence of endometrial cancer [1]. ICI is a relatively new ‘Type II’ antiestrogen [4]. Although ICI binds uterine ER protein with affinity similar to that of
E2 [5], it has little or no uterotrophic activity in rodents, pigs and women, and appears to antagonize E2 induction of endometrial growth [6–10]. We have identified several genes, including ER, PR, GAPDH, cyclophilin, and c-fos, that are up-regulated in sheep endometrium by a single physiological dose of E2 [11,12]. Previously, we showed that tamoxifen also up-regulates expression of these endometrial genes [11]. Here, we examined whether ICI acts as an agonist or antagonist of E2 effects on growth and gene expression in the uteri of ovariectomized sheep.
2. Materials and methods Abbre6iations: Con, control; E2, estradiol; ER, estrogen receptor-a; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; ICI, ICI 182,780; PR, progesterone receptor. * Corresponding author. Tel.: +1-979-8622790; fax: + 1-9798623399. E-mail address: [email protected] (N.H. Ing).
2.1. Animal treatments and sample collection Crossbred Rambouillet ewes (O6is aries) were ovariectomized during the breeding season after exhibiting breeding cycles of normal duration (16 –18
0960-0760/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S0960-0760(01)00061-9
282
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
days). Four weeks after ovariectomy, ewes were randomly assigned to four treatment groups (n = 4 ewes per group). Time of E2 injection (50 mg in charcoalstripped corn oil, intramuscular) was 18 h prior to hysterectomy, a time of rapid ER mRNA accumulation in endometrium [13]. ICI was administered for 14 h starting 24 h before hysterectomy (6 h prior to E2 treatment) and ending 10 h prior to hysterectomy (8 h subsequent to the E2 challenge) in order to competitively occupy ER protein. ICI (10 − 7 M ICI in 0.1% ovine serum albumin in phosphate-buffered saline) was delivered by infusing uteri via catheters seated in the tips of both uterine horns with either at a rate of 3 ml/h per horn. This rate is based on volumes known to be absorbed over time without causing uterine distension. At hysterectomy, uterine cross-sections were taken from the middle of one uterine horn just distal to the external bifurcation and fixed in 4% (w/w) paraformaldehyde before being embedded in Ameraffin (Baxter Diagnostics; Deerfield, IL). Endometrium was dissected from the remaining uterus, minced, snap-frozen in liquid nitrogen and stored at −80°C for RNA slot blot analyses. All animal procedures were approved by the Texas A&M University Laboratory Animal Care and Use Committee. Unless otherwise indicated, chemicals were purchased from Sigma (St Louis, MO).
2.2. Synthesis of cRNA probes For probing endometrial RNA immobilized on slot blots, cRNA probes were generated from ovine ER, PR, GAPDH, and c-fos cDNA as described previously [11,13]. The cRNA probes for cyclophilin and actin were generated from pTRI-cyclophilin-Human and pTRI-b-actin-Human (Ambion; Austin, TX). Briefly, antisense cRNA probes were generated by in vitro transcription with T7 RNA polymerase (Maxiscript kit; Ambion) and [32P]UTP (New England Nuclear; Boston, MA). Unincorporated radionucleotides were removed on spin columns (Bio 101; La Jolla, CA). For in situ hybridization, antisense and sense ER, PR, and GAPDH cRNA probes were produced as above except that [35S]UTP was substituted for [32P]UTP and the sense probes were synthesized with SP6 RNA polymerase.
2.3. Preparation of RNA samples and RNA slot blot analyses Total cellular RNA was purified from quadruplicate tissue samples (0.1 g each) with the Tripure Reagent (Roche Molecular Biochemicals; Indianapolis, IN). Absorbances at 260 nm were measured in duplicate and 5 mg total cellular RNA were loaded
on replicate slot blots with a BioRad vacuum manifold as previously described [11]. High stringency was maintained throughout blot hybridization (55°C) and washing {68°C in 0.1× SSC (1 × SSC = 150 mM NaCl + 15 mM Na citrate) and 0.1% sodium dodecyl sulfate followed by treatment with 1 mg/ml RNase A (Roche Molecular Biochemicals) for 10 min at room temperature}. Hybridization signals on blots were counted on an InstantImager (Packard; Meriden, CT). Prior Northern analyses using this protocol and these probes detected only bands of the expected sizes for each of the mRNAs analyzed [12].
2.4. In situ hybridization Serial 7-mm cross-sections from the uterus of each ewe were placed on Superfrost Plus slides (Curtin Matheson Scientific; Houston, TX). Within 1 week of sectioning, cross-sections were hybridized with one of six radiolabeled probes: sense and antisense ER, PR, and GAPDH cRNAs as described previously [14]. Slides were developed after 6 weeks of exposure to autoradiographic emulsion (NTB-2 from Kodak; Rochester, NY). Sections were counterstained lightly with hematoxylin. Densities of silver grains over different uterine cell types were quantitated on a Reichert Microstar IV microscope and NIH Image software version 1.61. Pixel densities were averaged on each uterine cross-section for each antisense probe over the following tissue compartments: luminal epithelium (LE), stratum compactum (SC, dense stroma just beneath the LE), superficial glandular epithelium (SGE), the adjacent superficial stroma (SS), deep glandular epithelium (DGE), deep stroma (DS), inner circular myometrium (Mi) and outer longitudinal myometrium (Mo) [14]. For each compartment, pixel densities on sections probed with sense cRNAs (negative controls for non-specific hybridization) were subtracted as background signals.
2.5. Statistical analyses Quantitative data were analyzed by least squares ANOVA using the General Linear Model procedures of SAS [15]. Results from treatment groups were compared: E2 versus Con for E2 effects, ICI versus Con for agonist effects of ICI, and ICI+ E2 versus E2 for antagonistic effects of ICI. Data are presented as least squares means and S.E. for treatment groups. Messenger RNA ‘concentrations’ from slot blots are expressed as a percentage of the average hybridization signal of the Con group. In situ hybridization signals are reported as pixel densities. Level of statistical significance is PB0.05 unless otherwise stated.
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
283
to the small, pale tracts of the Con ewes and the large, turgid, reddened tracts of the E2 ewes. Uteri from the ICI+ E2 treatment group were indistinguishable in outward appearance from those treated with E2 alone. Wet weights of dissected endometrium ranged from 5.7 to 12.7 g. The wet weights of endometrium were similar in ewes from E2, ICI and ICI + E2 groups and were an average of 50% greater compared to those of control ewes (Fig. 1). These results are consistent with ICI acting as an E2 agonist, not as an antagonist.
3.2. E2 regulates expression of endometrial genes
Fig. 1. Both E2 and ICI treatments increased dissected endometrial weights. Adult ewes were treated with E2, ICI or ICI +E2 as described in the text (n =4 ewes per treatment group). Wet weights of endometrium are reported in g as least squares means and S.E. for treatment groups. E2 and/or ICI treatments increased endometrial weights compared to controls (‘Con’) ewes (asterisks indicate differences of P =0.06).
3. Results
3.1. ICI affects gross uterine characteristics similar to E2 Whole uteri of ewes treated with ICI alone were intermediate in gross outward appearance compared
Gross endometrial levels of ER, PR, GAPDH, cyclophilin, actin and c-fos mRNAs were measured on slot blots. Quantitation of the hybridization signals indicated that E2 treatment up-regulated the expression of ER mRNA 70% relative to Con values (Fig. 2). E2 also up-regulated PR, GAPDH, cyclophilin and actin mRNA levels in endometrium with effects ranging from 37% (PR, PB 0.06) to 85% (actin). E2 had the opposite effect on c-fos mRNA: it down-regulated it to 38% of Con values. ICI, alone or in combination with E2, did not affect the concentrations of the mRNAs measured (P\ 0.4). However, antagonism was suggested by the fact that, for all five E2 up-regulated mRNAs, the hybridization signals in samples from ICI+ E2 ewes appeared to be an average of 12% lower than those from E2 treatment group ewes.
Fig. 2. Estradiol treatment affected mRNA levels in endometrium. Total cellular RNA isolated from quadruplicate samples of endometrium analyzed on slot blots for ER, PR, GAPDH, cyclophilin (‘Cyclophil’), actin and c-fos mRNAs. Results are presented as least squares means and S.E., expressed as percentage of the mean value of Con. E2 (filled bars) increased all mRNA concentrations, except it decreased that of c-fos (P5 0.06). ICI alone (hatched bars) or with E2 (stippled bars) had no significant effect, agonistic or antagonistic to that of E2.
284
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
Fig. 3. In situ hybridization of GAPDH mRNA indicates gene expression in specific uterine cell compartments. Representative views of in situ hybridization results on uterine cross-sections from ewes in Con, E2, ICI and ICI +E2 treatment groups are shown in panels A – D, respectively. Black silver grains on the brightfield images indicate in situ hybridization of the antisense GAPDH cRNA probe. Nuclei, lightly stained with hematoxylin, are visible primarily in panel A due to its lower hybridization signals. Panels span the upper (superficial) half of endometrium from luminal epithelium (LE) to superficial endometrial glands (SGE) extending down from the uterine lumen, at far right. The bar represents 100 mm.
3.3. In situ hybridization analyses show cell-specific effects of ICI To examine effects at the cellular level, ER, PR and GAPDH cRNA probes were used for in situ hybridization. The up-regulation of ER and PR mRNAs have been extensively analyzed previously [14]. However, this is the first in situ characterization of GAPDH mRNA in sheep uteri (Fig. 3). It demonstrated that GAPDH mRNA was more strongly expressed in epithelial cells of the endometrium compared to endometrial stroma and myometrium. The low expression levels of GAPDH mRNA in Con ewes (Fig. 3A) were strongly up-regulated by E2 in the stromal cells (SC and SS) below the epithelial cells lining the uterine lumen (LE), as well as within the epithelium of the upper glands (SGE; Fig. 3B). ICI acted as an E2 agonist in up-regulating GAPDH mRNA in SC and SGE cell types (Fig. 3C) and did not inhibit the E2 up-regulation described above (Fig. 3D). Hybridization signals for ER, PR and GAPDH mRNAs were quantitated over cell compartments spanning the endometrium and myometrium. The strengths of the hybridization signals were, in general, strongest for ER mRNA and weakest for PR mRNA (Fig. 4). For all three mRNAs analyzed, expression was lowest in the
less dense stromal compartments (SS and DS). The only significant effect of E2 was to increase ER mRNA 207% in Mi (Fig. 4, top panel). However, all compartments showed a trend in increasing ER mRNA in response to E2. ICI treatment, when used alone, had no effect. When used in conjunction with E2, however, ICI inhibited the rise in ER mRNA in Mi. Results of PR mRNA in situ hybridization were similar to those for ER mRNA: E2 up-regulated PR mRNA 221% in the Mi cell compartment and ICI inhibited the up-regulation (Fig. 4, middle panel). Quantitation of the in situ hybridization of GAPDH mRNA demonstrated that E2 increased GAPDH mRNA in SC, SGE, SS, DGE and Mi compartments between 142 and 442% (Fig. 4, bottom panel). ICI treatment mimicked that of E2 in up-regulating GAPDH mRNA in SC and SGE cells 208 and 142%, respectively. ICI did not antagonize E2 effects on GAPDH gene expression.
4. Discussion The uterus, and especially the endometrium, is exquisitely responsive to E2. We showed previously [11] that a single 50-mg dose of E2 increased uterine wet
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
weight, a classic bioassay for estrogenic effects. In this study, ICI acted like E2 by increasing uterine redness and endometrial weight. ICI also acted as an estrogen agonist on uterine parameters involved in luteolysis in the only other published study employing adult sheep [16]. Uterine gland growth, responsive to E2 treatment
285
in neonatal lambs, pigs and calves ([8], and references therein), also increased in neonatal lambs with ICI treatment (125 mg/kg per day for 14 days; Dr Thomas E. Spencer, pers. commun.). The same ICI treatment of neonatal pigs had no effect on uterine gland growth [8]. The fact that three studies with sheep have found
Fig. 4. Quantitation of in situ hybridization results for ER, PR and GAPDH mRNAs indicates changes in gene expression in specific uterine cell compartments. Pixel densities measured by light microscopy and NIH Image software are reported for ER, PR and GAPDH (top, middle, and bottom panels, respectively). The different uterine cell compartments analyzed are: luminal epithelium (LE), stratum compactum (SC, dense stroma just beneath the LE), superficial glandular epithelium (SGE), the adjacent superficial stroma (SS), deep glandular epithelium (DGE), deep stroma (DS), inner circular myometrium (Mi) and outer longitudinal myometrium (Mo). Least squares means and S.E. of pixel densities on uterine cross-sections from Con, E2, ICI and ICI + E2 treatment groups are reported in open, filled, hatched and stippled bars, respectively. Asterisks (*) and number symbols (c ) indicate differences compared to Con and E2 groups, respectively.
286
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
uterotropic actions of ICI may indicate a true species difference of sheep with the rat, pig, monkey, and human species, in which ICI acts as a pure estrogen antagonist [4,6–10,13]. The amino acid sequences of sheep, pig and human ER proteins (GenBank accession numbers Z49257, Z37167 and X03635, respectively) are highly conserved: 95% (sheep/pig) and 91% (sheep/human) identical overall. However, the C-terminus is much less conserved: 72% (sheep/pig) and 63% (sheep/ human) identical over the last 43 residues. Since the C-terminus of ER affects ligand discrimination [17], perhaps these differences alter the effect of ICI with the sheep ER compared to ICI with the pig and human ER proteins. We examined ICI effects, with and without E2, on a battery of genes known to respond to E2 in the sheep endometrium [11,12]. Consistent with previous results, E2 increased gross endometrial expression of ER, PR, GAPDH, and cyclophilin genes. Additionally, we report for the first time that E2 up-regulates actin mRNA in sheep endometrium. Although many regard GAPDH, cyclophilin and actin genes as unresponsive, ‘house-keeping’ genes, others have reported their upregulation by E2 in the rodent uterus [18,19]. Thus, care should be taken in normalizing data to expression of a ‘house-keeping’ gene, especially during a vigorous response like that of the uterus under the influence of E2. In this study, E2 down-regulated c-fos mRNA. Our other studies in ovariectomized ewes have shown early c-fos up-regulation [11] and later down-regulation [12]. This biphasic response of the c-fos gene has been explained by early induction of transcription by activated ER protein followed by autologous down-regulation by the c-fos protein itself [20]. Tamoxifen, a predecessor of ICI, acted like E2 in up-regulating ER and GAPDH gene products in sheep endometrium [11]. In this study, ICI did not act as an agonist or antagonist on gross levels of any of the E2-regulated gene products. However, in situ hybridization allowed dissection of the separate uterine cell compartments for individual assessment of their changes in gene expression in response to E2 and ICI. With it, we are the first to report that the GAPDH gene is expressed predominantly in the endometrial epithelium and myometrial cells of the sheep uterus. E2 up-regulated GAPDH mRNA levels in most uterine cell types. In addition, ICI acted like E2 in up-regulating GAPDH mRNA in the superficial stroma and glandular epithelium of the endometrium. Other examples of ICI acting as an estrogen agonist have been described on a subset of estrogen-responsive gene promoters in uterine cell lines and primary cultures [21– 23]. There is also evidence that ICI can activate these promoters with estrogen receptor-b. However, estrogen receptor-b gene expression in the uterus is very low relative to that of ER-a [24,25].
Regulation of ER and PR genes by E2 in specific uterine cells was characterized previously [14]. This report is consistent with that study: E2 markedly enhanced ER and PR mRNA levels in the innermost layer of myometrium. In this robustly responsive uterine compartment, the in situ hybridization data revealed that ICI antagonized the E2 up-regulation of both ER and PR mRNAs. These data agree with the majority of studies employing ICI as an estrogen antagonist in vivo. For example, ICI antagonized the estrogen induction of PR, c-fos, lactoferrin and vascular endothelial growth factor genes in mouse uterus [26,27]. Those studies used ratios of ICI to E2 ranging from 25 to 100, which brackets our estimated ratio of 70 (based on the assumption that the endometrium equilibrates to 10 − 7 M ICI in the infusate and retains 850 pg E2 per g ($ ml) endometrium from the injection [11]). The fact that the E2 antagonism occurred in myometrium, the furthest tissue from the luminal delivery site of the ICI, confirms that the inhibitor delivered in the 42 ml of total infusate penetrated the entire uterus (25–40 g wet wt. [11]). One explanation of ICI antagonism of E2 effects is that ICI down-regulates levels of ER protein [1]. To see if ER protein levels were preferentially changed in the cell compartments that responded to ICI, we performed immunohistochemistry with the anti-ER H222 antibody on uterine cross-sections. Although, we saw no differences in endometrial or myometrial ER protein levels with ICI treatment (data not shown), it is possible that prior down-regulation of ER protein levels recovered by the time of hysterectomy. However, some reported increases in uterine ER protein during negative effects on uterine growth in mice treated with ICI [28]. It is important to note that estrogens rapidly down-regulate levels of immunoreactive ER protein in cells [13,29] so they may not directly correlate temporally to the action of E2 in a cell, especially during acute treatments. Rather, specific cells may activate cofactors, corepressors, and/or convergent signaling pathways differentially in response to E2 and/or ICI [21,30]. We conclude that ICI effects on estrogen-regulated gene expression, a balance between agonism and antagonism, are strongly dependent upon the species, gene, and cell type being studied [1,11,30].
Acknowledgements The authors gratefully acknowledge the generous gift of ICI 182,780 from Dr Alan Wakeling of Zeneca Pharmaceuticals, without which the study would not have been possible. Also instrumental were the technical assistance of Cindy Balog and the Image Analysis core facility of Dr Robert C. Burghardt and the Center for Environmental and Rural Health. Research was
J.A. Robertson et al. / Journal of Steroid Biochemistry & Molecular Biology 77 (2001) 281–287
conducted by the Texas Agricultural Experiment Station, with funding provided by award c 95-37203-2182 to NHI from the USDA NRI-CGP.
References [1] J.I. MacGregor, V.C. Jordan, Basic guide to mechanisms of antiestrogen action, Pharm. Rev. 50 (1998) 151 – 196. [2] P.G. Satyaswaroop, Estrogenic and antiestrogenic actions of tamoxifen in the female reproductive tract, in: R. Lindsay, D.W. Dempster, V.C. Jordan (Eds.), Estrogens and Antiestrogens, Lippincott-Raven, Philadelphia, PA, 1997, pp. 2121 –2221. [3] C.L. Lin, H.L. Buttle, Effect of oestradiol benzoate and tamoxifen on the growth and induction of progesterone receptors in the uterus and mammary gland of immature pigs, J. Endocrinol. 130 (1991) 259 –265. [4] A.E. Wakeling, M. Dukes, J. Bowler, A potent specific pure antiestrogen with clinical potential, Cancer Res. 51 (1991) 3867 – 3873. [5] C. Martel, L. Provencher, X. Li, A. St. Pierre, G. Leblanc, S. Gauthier, Y. Merand, F. Labrie, Binding characteristics of novel nonsteroidal antiestrogens to the rat uterine estrogen receptors, J. Steroid Biochem. Mol. Biol. 64 (1998) 199 –205. [6] W.S. Branham, R. Fishman, R.D. Streck, K.L. Medlock, J.J. DeGeorge, D.M. Sheehan, ICI 182,780 inhibits endogenous estrogen-dependent rat uterine growth and tamoxifen-induced developmental toxicity, Biol. Reprod. 54 (1996) 160 –167. [7] L. Sahlin, H. Eriksson, Influence of blockers for the estrogen receptor (ER) and type 1 IGF-receptor on the levels of ER mRNA and IGF-1 mRNA in the rat uterus, J. Steroid Biochem. Mol. Biol. 58 (1996) 359 –365. [8] B.J. Tarleton, A.A. Wiley, F.F. Bartol, Endometrial development and adenogenesis in the neonatal pig: effects of estradiol valerate and the antiestrogen ICI 182,780, Biol. Reprod. 61 (1999) 253 –263. [9] M. Dukes, J.C. Waterton, A.E. Wakeling, Antiuterotrophic effects of the pure antioestrogen ICI 182,780 in adult female monkeys (Macaca nemestrina): quantitative magnetic resonance imaging, J. Endocrinol. 138 (1993) 203 – 209. [10] E.J. Thomas, P.L. Walton, N.M. Thomas, M. Dowsett, The effects of ICI 182,780, a pure anti-oestrogen, on the hypothalamic– pituitary –gonadal axis and on endometrial proliferation in pre-menopausal women, Hum. Reprod. 9 (1996) 1991 –1996. [11] J.A. Robertson, S. Bhattacharyya, N.H. Ing, Tamoxifen up-regulates oestrogen receptor-alpha, c-fos and glyceraldehyde 3phosphate-dehydrogenase mRNAs in ovine endometrium, J. Steroid Biochem. Mol. Biol. 67 (1999) 285 –292. [12] K. Zou, N.H. Ing, Oestradiol up-regulates oestrogen receptor, cyclophilin, and glyceraldehyde phosphate dehydrogenase mRNA concentrations in endometrium but down-regulates them in liver, J. Steroid Biochem. Mol. Biol. 64 (1998) 231 –237. [13] N.H. Ing, T.E. Spencer, F.W. Bazer, Estrogen enhances endometrial estrogen receptor gene expression by a posttranscriptional mechanism in the ovariectomized ewe, Biol. Reprod. 54 (1996) 591 – 599.
287
[14] N.H. Ing, M.B. Tornesi, Estradiol up-regulates estrogen receptor and progesterone receptor gene expression in specific ovine uterine cells, Biol. Reprod. 56 (1997) 1205 – 1215. [15] SAS User’s Guide: Basics. SAS Institute, Cary, 1982. [16] H. Al-Matubsi, R.J. Fairclough, G. Jenkin, Oestrogenic effects of ICI 182,780, a putative anti-oestrogen on the secretion of oxytocin and prostaglandin F2a during oestrous cycle in the intact ewe, Anim. Reprod. Sci. 51 (1998) 81 – 96. [17] B.S. Katzenellenbogen, M.M. Montano, P. Le Goff, D.J. Schodin, W.L. Kraus, B. Bhardwaj, N. Fujimoto, Antiestrogens: mechanisms and actions in target cells, J. Steroid Biochem. Mol. Biol. 53 (1995) 387 – 393. [18] M. Rosser, L. Chorich, E. Howard, P. Zamorano, V.B. Mahesh, Changes in rat uterine estrogen receptor messenger ribonucleic acid levels during estrogen- and progesterone-induced estrogen receptor depletion and subsequent replenishment, Biol. Reprod. 48 (1993) 89 – 98. [19] L. Sahlin, Dexamethasone attenuates the oestradiol-induced increase in IGF-1 mRNA in the rat uterus, J. Steroid Biochem. Mol. Biol. 55 (1995) 9 – 15. [20] M. Schuchard, J.P. Landers, N.P. Sandhu, T.C. Spelsberg, Steroid hormone regulation of nuclear proto-oncogenes, Endocr. Rev. 14 (1993) 659 – 669. [21] E. Castro-Rivera, S. Safe, Estrogen and antiestrogen-responsiveness of HEC1A endometrial adenocarcinoma cell in culture, J. Steroid Biochem. Mol. Biol. 64 (1998) 287 – 295. [22] P.S. Jones, E. Parrott, I.N.H. White, Activation of transcription by estrogen receptor a and b is cell-type and promoter-dependent, J. Steroid Biochem. Mol. Biol. 274 (1999) 32008 –32014. [23] K. Paech, P. Webb, G.G.J.M. Kuiper, S. Nilsson, J.-A. Gustafsson, P.J. Kushner, T.S. Scanlan, Differential ligand activation of estrogen receptors ERa and ERb at AP1 sites, Science 277 (1997) 1508 – 1510. [24] G.G.J.M. Kuiper, B. Carlsson, K. Grandien, E. Enmark, J. Haggblad, S. Nilsson, J.-A. Gustafsson, Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptor a and b, Endocrinology 138 (1997) 863 –870. [25] H. Wong, B. Masironi, H. Eriksson, L. Sahlin, A comparative study of estrogen receptors a and b in the rat uterus, Biol. Reprod. 62 (2000) 310 – 317. [26] S.K. Das, J. Tan, D.C. Johnson, S.K. Dey, Differential spatiotemporal regulation of lactoferrin and progesterone receptor gene in the mouse uterus by primary estrogen, catechol estrogen, and xenoestrogen, Endocrinology 139 (1998) 2905 – 2915. [27] S.M. Hyder, C. Chiapetta, L. Murthy, G.M. Stancel, Selective inhibition of estrogen-regulated gene expression in vivo by the pure antiestrogen ICI 182,780, Cancer Res. 57 (1997) 2547 – 2549. [28] C. Martel, C. Labrie, A. Belanger, S. Gauthier, Y. Merand, X. Li, L. Provencher, B. Candas, F. Labrie, Comparison of the effects of the new orally active antiestrogen EM-800 with ICI 182 780 and toremifene on estrogen-sensitive parameters in the ovariectomized mouse, Endocrinology 139 (1998) 2486 –2492. [29] Z. Nawaz, D.M. Lonard, A.P. Dennis, C.L. Smith, B.W. O’Malley, Proteasome-dependent degradation of the human estrogen receptor, Proc. Natl. Acad. Sci. USA 96 (1999) 1858 – 1862. [30] B.S. Katzenellenbogen, M.M. Montano, K. Ekana, M.E. Herman, E.M. McInerney, Antiestrogens: mechanisms of action and resistance in breast cancer, Breast Cancer Res. Treat. 44 (1997) 23 – 38.