Search

The aim of the study was to determine whether the dihydroxylated antiestrogen LY117018, with a high affinity for the estrogen receptor and low intrinsic estrogenic activity, could inhibit the uterotropic actions of steroidal [estradiol-17 beta (E2)] and nonsteroidal [ICI 3188 and trianisylchloroethylene (TACE)] estrogens in immature rats and also the uterotropic actions of tamoxifen and monohydroxytamoxifen in the ovariectomized mouse and immature rat. In the first series of experiments, LY117018 was compared with monohydroxytamoxifen. Both antiestrogens inhibited the uterotropic actions of E2 (0.32 micrograms daily), ICI 3188 (5 micrograms daily), and TACE (40 and 160 micrograms daily) in a dose-related manner (0.32-82 micrograms daily). The potency of the antiestrogens against E2 and ICI 3188 was similar, however, at higher doses (20.48 and 82 micrograms daily) LY117018 reduced uterine weights to below the lowest level achieved by monohydroxytamoxifen. In contrast, LY117018 was less effective against the long acting estrogen TACE. The competitive interaction of LY117018 with tamoxifen and monohydroxytamoxifen was compared in 3-day ovariectomized mouse and immature rat uterine weight tests. Tamoxifen and monohydroxytamoxifen were fully estrogenic in the mouse (5 micrograms daily) and partially estrogenic in the rat (1.5-20 micrograms daily). LY117018 was a partial estrogen in the mouse and a weekly active partial estrogen in the rat (2.5-120 micrograms daily). LY117018 produced dose-related decreases in the uterine weight increases induced by tamoxifen and monohydroxytamoxifen in both species. However, in the rat, LY117018 was more effective against the less potent compound tamoxifen, at a 6:1 dosage ratio compared with a 24:1 dosage ratio required for the potent compound monohydroxytamoxifen. LY117018 had a short duration of action as an antiestrogen when compared with monohydroxytamoxifen. LY117018 (120 micrograms) was only completely effective as an antiestrogen if administered repeatedly with E2 whereas a single injection of monohydroxytamoxifen (120 micrograms) was sufficient to inhibit fully E2 action in the uterus for up to 4 days. Because LY117018 has a shorter duration of action than monohydroxytamoxifen, a high dosage ratio of LY117018 over monohydroxytamoxifen is required to maintain effective competitive antagonism in the uterus. Overall, these findings suggest that monohydroxytamoxifen and LY117018 probably act through the same mechanism of action via the estrogen receptor.

The nonsteroidal antiestrogen tamoxifen, 4-hydroxytamoxifen, and Ly117018 inhibited the estradiol-stimulated induction of progesterone receptors in primary cultures of immature rat uterine cells. This effect was found to be completely reversible with increased concentrations of estradiol. These compounds possessed no estrogenic activity. In contrast, ICI 47,699 (the cis geometric isomer of tamoxifen) and ICI 77,949 (tamoxifen without the dimethylaminoethyl side chain) were fully estrogenic, and bisphenol (4-hydroxytamoxifen without the dimethylaminoethyl side chain) possessed mixed estrogenic/antiestrogenic activity. In primary uterine cell cultures derived from mature ovariectomized mice, 4-hydroxytamoxifen was, again, nonestrogenic and inhibited the estradiol-stimulated induction of progesterone receptors. The antiestrogenic activity of 4-hydroxytamoxifen was effective against both steroidal and nonsteroidal estrogens in either rat- or mouse-derived uterine cell cultures. Using the 3-day uterine assay in vivo, 4-hydroxytamoxifen partially stimulated progesterone receptor induction in the immature rat, whereas it fully stimulated the same end point in the mature ovariectomized mouse. These results emphasize the difference between antiestrogen activity in vivo and in vitro, and also indicate that the increased agonist activity of 4-hydroxytamoxifen in the mouse compared to that in the rat in vivo is not reflected in vitro. Therefore, we have extended the model of antiestrogen action previously described in primary pituitary cell cultures to progesterone receptor induction in two murine uterine cell cultures.

The estrogenic and antiestrogenic activities of derivatives of estradiol and estrone were determined in vitro using the ability of primary cultures of immature rat pituitary cells to synthesize PRL. Estradiol derivatives were the most potent estrogens in the assay. Large ethinyl substitutions in the 17 alpha position generally caused a decrease in estrogenic potency (up to 1000-fold). The 3 phenolic hydroxyl was important, but not essential, for the estrogenic activity of the estradiol molecule. Estratriene was approximately 1000 times less potent than estradiol. However, significant estrogenic activity was observed with the compound anordin (EC50, 8 x 10(-9) M), which could potentially be converted to a dihydroxylated derivative but without an aromatic A ring. Similarly, the steroid androst-5-ene-3,17-diol was weakly estrogenic (EC50, 3 x 10(-8) M). Steriods with a ketone in the A and D rings were generally inactive as estrogens and antiestrogens. Estradiol derivatives with 17 beta amines were only weak estrogens. Estrone derivatives were less active than the corresponding estradiol derivatives. 4-Nitromethoxyestrone exhibited weak antiestrogenic properties; however, 4-nitroestrone and methoxyestrone were both estrogens. The reason for the antiestrogenic properties of 4-nitromethoxyestrone is obscure, as the compound does not have structural features similar to those of known nonsteroidal antiestrogens. Minor alterations to the estradiol molecule at the 11 beta (OH) or 6 (ketone) position had little effect on estrogenic potency; however, large substitutions at the 11 beta (RU 39,411) or 7 alpha (ICI 164384) position produced antiestrogenic compounds. RU 39,411 was approximately 10 times more active as an antiestrogen than 4-hydroxytamoxifen, whereas ICI 164,384 was approximately 10 times less active than 4-hydroxytamoxifen. A series of hypothetical models is proposed that could explain the antiestrogenic properties of RU 39,411 and ICI 164,384 by an interaction with the estrogen receptor steroid-binding site.

The rat uterus responds to estradiol (E2) and E2 benzoate stimulation with an increase in progesterone receptor production and with growth. These responses were also elicited to varying degrees by a series of estrogenic (ICI 77,949 and ICI 47,699) and antiestrogenic triphenylethylene derivatives [tamoxifen (TAM), 4-hydroxy-TAM (4-OH-TAM), and 4-CH3-TAM]. These compounds have a range of affinities for the estrogen receptor (ER) and are able to compete with [3H]E2 binding in the uterus in vivo. Within 1-2 h of a sc injection of high affinity ligands (E2, E2B, and 4-OH-TAM), there was decrease in cytosol ER. This decrease was also observed with TAM, which is metabolized to 4-OH-TAM in vivo. In contrast, there was no decrease in cytosolic ER in animals treated with low affinity compounds (ICI 77,949, ICI 47,699, and 4-CH3-TAM) at any time before the onset of an estrogenic response. Furthermore, the nuclear ER increased after administration of a high affinity ligand (E2), as measured by exchange assay, but no increase in nuclear ER was observed after administration of low affinity ligands (ICI 77,949 and 4-CH3-TAM), although estrogenic responses were produced. From these data we have suggested a functional model to explain ER-mediated events in the rat uterus that supports the recent proposal that unoccupied ER is located in the nuclear compartment. In this model, the majority of unoccupied ER may reside in the nucleus in vivo; however, when the cells are disrupted in vitro, the unoccupied receptor or dissociation of low affinity ligand-ER complex causes unoccupied receptor to fall out of the nucleus and be incorporated into the cytosolic fraction. The high affinity ligand-ER complexes are retained in the nucleus. This would suggest that the apparent translocation of the ER from the cytoplasm to the nucleus may be an artifact. The data may reflect differential extraction of unoccupied receptors from the nucleus rather than transfer of receptor complexes to the nucleus.

The estrogenic and antiestrogenic actions of the geometric isomers of tamoxifen and 4-hydroxytamoxifen were determined in a PRL synthesis assay using primary cultures of dispersed immature rat pituitary gland cells. 4-Hydroxytamoxifen was 100 times more potent as an antiestrogen than tamoxifen. The cis isomer of tamoxifen was a weak estrogen, but the cis isomer of 4-hydroxytamoxifen was converted to the trans isomer during the 6-day assay. This made an accurate determination of the properties of cis-(E)4-hydroxytamoxifen impossible. A series of fixed ring derivatives of the compounds were evaluated in the PRL synthesis assay in vitro to determine their estrogenic and antiestrogenic activities. The fixed ring derivatives of tamoxifen, cis-tamoxifen and trans-(Z)4-hydroxytamoxifen all had properties that were the same as those of the original triphenylethylenes. The fixed ring derivative of the cis-(E) isomer of 4-hydroxytamoxifen was a weak competitive antagonist of estrogen action with only very slight estrogenic properties. This contrasted with the other cis isomers of triphenylethylenes. We propose that the hydroxyl group on the molecule may orient the ligand at the binding site of the estrogen receptor to place the alkylaminoethoxy side-chain in a position to produce antiestrogen action.

The intracellular location of estrogen receptors in hormone-responsive cells has been studied with a number of techniques which indicate that the unoccupied receptors are nuclear and not cytoplasmic proteins. We used cell enucleation of two human breast cancer-derived cell lines, MCF-7 and T47D, to determine whether the unoccupied receptors were also nuclear in these cells and to determine whether the weak estrogen phenol red, present in nearly all tissue culture media, affected the distribution of the receptors seen with this technique. Nucleoplasts prepared from the breast cancer cells contained most of the estrogen receptors that were present in whole cells. The cytoplast fraction, which contained some contaminating whole cells, also contained some receptors. However, incubating cells with estradiol before enucleation did not translocate any receptors out of the cytoplast fraction (to the nucleoplasts). The unoccupied receptors appeared to be almost exclusively nuclear in these cells. The same results were obtained with either radioligand binding or enzyme-linked immunoassay used to measure estrogen receptor, and the distribution of receptors was unaffected by the presence of the pH indicator phenol red. In addition, we observed changes in the estrogen receptor content of incubated cytoplasts that were consistent with receptor synthesis, and this may prove to be a useful model system to characterize receptor synthesis and degradation.