Thomas S. Ruh

Antiestrogen action: Differential nuclear retention and extractability of the estrogen receptor

Since there is a much longer uterine nuclear retention of the U-11, 100A (antiestrogen) receptor complex (UARC) than of the estradiol receptor complex (ERC) at 4-12 hrs after injection, experiments were designed to determine if there is a difference between the relative nuclear affinities for the two RCs as determined by extraction with various ionic strength mediums. Although the UARC was retained longer in the nuclear fraction in vivo, the UARC was completely extractable with 0.3M KCl or 50mM spermine, whereas the ERC demonstrates a salt-resistant form. This suggests that the ERC is more tightly bound to nuclear components through this salt-resistant form of the receptor. In addition, various intercalating agents were used to distinguish the different nuclear chromatin DNA sites where the UARC and ERC may be binding. With actinomycin D (50 uM) more ERC than UARC was retained in the nuclear fraction. However, with ethidium bromide (100uM) less ERC than UARC was retained. Also, the ERC selectively released by ethidium bromide is precisely that fraction not released by salt. These results indicate that the UARC and ERC bind to different chromatin loci.

THE EFFECTS OF ANTIOESTROGENS ON THE OESTROGEN RECEPTOR

Abstract In order to probe the source of the antagonistic properties of triphenylethylene antioestrogens in oestrogen induced uterine growth, we have studied both low- and high-affinity antioestrogens. In in vitro competition studies H-1285 was shown to have about 1100% the affinity for the oestrogen receptor when compared with 17-β-oestradiol (E2) (100%) which is in sharp contrast to H-1067 and U-11, 100A (UA) which have 25 and 4% binding affinities, respectively. H-1285 and H-1067 both caused an increase in uterine DNA and wet weight, but were impeded oestrogens. H-1285 and H-1067 were also antioestrogenic because they inhibited the E2-induced DNA and wet weight responses. The in vivo nuclear and cytoplasmic oestrogen receptor responses to UA, H-1067 and H-1285 were also investigated and compared to the E2 response. All antioestrogens studied caused a prolonged nuclear retention of the oestrogen receptor, in contrast to the rapid uptake followed by a rapid deletion seen with E2. H-1285 receptor complexes, however, displayed a delayed uptake, which did not reach maximal nuclear levels until 24 h. Cytoplasmic oestrogen receptor replenishment with E2 was immediate and rapid. All the antioestrogens studied demonstrated delayed and slow receptor replenishment, about 25% the E2 induced rate. E2 injected at 48 h subsequent to an injection of H-1285 caused a rapid 1-h nuclear translocation of the oestrogen receptor followed by a complete nuclear “fallout” of receptor, a phenomenon previously seen only with oestrogens. No salt-resistant oestrogen receptor complex was measurable with a high-affinity antioestrogen H-1285 or with low affinity antioestrogens, H-1067, UA or enclomiphene during 1–12 h, but the salt-resistant form was seen with E2 and diethylstilboestrol at 1–6 h. All of the oestrogen responses (wet weight, protein content, DNA and protein synthesis) exhibited the same increase at 24 h whether single 5 μg E2 or H-1285 injections were administered. However, by 48 and 72 h, H-1285 consistently gave higher responses than did E2. E2 administered at 48 h subsequent to a previous E2 injection caused a dramatic increase in oestrogenic responses, in contrast to E2 given subsequent to H-1285 which caused only a slight increase in these responses. These data demonstrate the ability of H-1285 to inhibit oestrogen induced uterine responses.

Binding of [3H]estradiol- and [3H]H1285-receptor complexes to rabbit uterine chromatin.

In the present study we investigated the binding characteristics of estrogen and antiestrogen-receptor complexes to rabbit uterine chromatin. Activated or nonactivated estrogen receptors were partially purified by DEAE-cellulose chromatography using low (1 mM) or high (10 mM) concentrations of sodium molybdate. Activated [3H]estradiol-receptor complexes showed enhanced binding to chromatin acceptor sites unmasked by 1 M, 4 M and 6 M guanidine hydrochloride. We also examined the chromatin-binding characteristics of the estrogen receptors when bound by the high-affinity triphenylethylene antiestrogen, H1285. The acceptor site activity for the [3H]H1285-receptor complexes was markedly decreased at sites unmasked by 4 M and 6 M guanidine hydrochloride. Further, the nonactivated receptor complexes showed very low binding to deproteinized chromatin. The estrogen-receptor chromatin-acceptor sites were tissue specific and saturable. These chromatin acceptor sites differ in their affinity and capacity (number of binding sites per cell) for the estrogen- and antiestrogen-receptor complexes. Thus, we suggest that the differences in the physiological and physiochemical properties of estrogens and antiestrogens may be related to their differential interaction with uterine chromatin subfractions.

Differences in the form of the salt-transformed estrogen receptor when bound by estrogen versus antiestrogen

Our laboratory has previously reported that antiestrogen binding to molybdate-stabilized non-transformed estrogen receptor results in a larger form of the receptor in 0.3 M KCl when compared with estrogen bound receptor. Estradiol promoted the formation of monomers in the presence of 0.3 M KCl whereas antiestrogen appeared to promote dimer formation. We have extended these studies examining the rabbit uterine salt-transformed estrogen receptor partially purified by DEAE-cellulose chromatography. We previously demonstrated that estrogen receptor prepared in this way bound to different sites on partially deproteinized chromatin subfractions or reconstituted chromosomal protein/DNA fractions when the receptor was complexed with estrogen vs antiestrogen. Analysis of these receptor preparations indicated that DEAE-cellulose step-elution resulted in a peak fraction which sedimented as a single 5.9S peak in 5-20% sucrose density gradients containing 0.3 M KCl for receptor bound by the antiestrogens H1285 and trans-hydroxytamoxifen. However, receptor bound by estradiol sedimented as 4.5S. These receptor complexes bound DNA-cellulose indicating that these partially purified receptors were transformed. DEAE rechromatography or agarose gel filtration of the partially purified antiestrogen-receptor complexes resulted in significant dissociation of the larger complex into monomers. Incubations of 5.9S antiestrogen-receptor complexes with antibodies against nontransformed steroid receptor-associated proteins (the 59 and 90 kDa proteins) did not result in the interaction of this larger antiestrogen-receptor complex with these antibodies (obtained from L. E. Faber and D. O. Toft, respectively). Our results support the concept that antiestrogen binding induces a different receptor conformation which could affect monomer-dimer equilibrium, thus rendering the antiestrogen-receptor complex incapable of inducing complete estrogenic responses in target tissues.

Binding of the estradiol-receptor complex to reconstituted nucleoacidic protein from calf uterus

Non-histone protein-DNA complexes with acceptor activity for estradiol-receptor complexes were reconstituted from fractionated calf uterine chromatin. Acceptor activity had tissue specificity with target tissue binding exceeding non-target tissue binding. The binding of estradiol-receptor complexes to acceptor sites was dependent on intact non-histone protein-DNA complexes, reconstituted select non-histone proteins, and protein equivalent: DNA reconstitution ratios. [3H]Estradiol-receptor complexes were bound to reconstituted non-histone protein-DNA complexes (i.e., nucleoacidic protein) with a high affinity and with a limited number of binding sites. Fractionation of uterine chromatin non-histone proteins identified two major sets of non-histone proteins which had acceptor activity when reconstituted with DNA. Thus, it seems possible to reconstitute nucleoacidic protein fractions with specific acceptor activity for the calf uterine estrogen receptor.

Nuclear acceptor sites for steroid hormone receptors: Comparisons of steroids and antisteroids

The respective chromatin binding sites (acceptor sites) for both the avian oviduct progesterone receptor (PR) and the rabbit uterine estrogen receptor (ER) reported by two separate laboratories are compared. Support for a saturable, high affinity binding to the chromatin acceptor sites by both receptors is described. Nonradiolabelled PR or ER compete with their homologous radiolabelled receptors for binding. However, there is no competition between the heterologous ER and PR for the nuclear binding sites. In both receptor systems, evidence for a receptor dependent, receptor specific binding, which mimics the binding measured in vivo, is reported. With both the PR and ER, evidence for extensive masking in chromatin of many of the acceptor sites is described. The PR and ER acceptor sites appear to be composed of specific acceptor proteins bound to DNA. The dissociation of these proteins from DNA causes a loss of specific binding. Reannealing of these same chromatin protein fractions, but not other protein fractions, back to the DNA reinstates the specific PR binding. Antibodies against the PR acceptors proteins are described. These antibodies block PR but not ER binding to the chromatin acceptor sites, supporting the steroid receptor specific acceptor sites. In the rabbit uterine system, the ER acceptor proteins dissociate from chromatin in three distinct fractions according to their affinity for the DNA. Each of these fractions contains acceptor activity in that each can be reannealed to the DNA to reconstitute specific ER binding sites, i.e. those which are saturable, high affinity, and receptor dependent. Interestingly, the antiestrogen receptor complexes do not bind to one fraction of acceptor proteins which does bind the native estrogen receptor complex. This ER acceptor protein fraction contains two regions of acceptor activity in the molecular mass range of 50,000 and 12,000 daltons. The latter species is in the size range of the PR acceptor proteins. This difference in binding to chromatin acceptor sites by the estrogen receptor, complexed with an estrogen or antiestrogen, may explain the differential regulation of gene expression and the differential biological response to antiestrogens compared to the native estrogens.

The effects of histone acetylation on estrogen responsiveness in MCF-7 cells

Because histone acetylation is implicated in the facilitation of specific gene transcription, the effect of increasing histone acetylation on the expression of an endogenous gene was investigated. The ability of trichostatin A (TSA), a histone deacetylase inhibitor, to potentiate the estradiol (E2) induction of progesterone receptor (PR) levels in MCF-7 cells was studied. Although TSA alone had no effect on PR synthesis, measured by a whole-cell binding assay with [3H]R5020, TSA potentiated the effect of 10−11M E2 such that 10 ng of TSA/mL approximately doubled the hormone response. When TSA was removed from the cells after various incubation times (24 and 48 h) by successive washings with TSA-free medium, it was determined that TSA was required throughout the 96-h incubation period in order to achieve maximal potentiation for the E2 response. In addition, TSA potentiated E2 induction of pS2 mRNA. These results suggested that the estrogen receptor (ER) complex might alter histone acetylation as part of the gene activation process. To test this directly, MCF-7 cells were incubated for 48 h with E2 followed by incubation with sodium [3H]acetate for 1 h. From two-dimensional polyacrylamide gel electrophoresis, an increase in total acetate incorporation into histones in estrogen- treated cells compared to control was observed as well as a preferential increase in the mono-and diacetylated histone H4. Experiments with lysinespecific antiacetylated H4 antibodies suggest a preferential increase in acetylation at lysine 16, but not 5, 8, or 12. The results of this study support an important role for histone acetylation in the mechanism of action of the ER.

Estrogen receptor interaction with specific histones: Binding to genomic DNA and an estrogen response element

Chromosomal proteins that impart high affinity and specificity to the binding of the estrogen receptor (ER) to DNA are termed estrogen receptor binding factors (ERBFs). Certain partially purified chromosomal protein fractions obtained from rabbit uterine chromatin by extraction with various molarities of GdnHCl when reconstituted to double-stranded DNA demonstrated high affinity binding for the ER. We report the purification and characterization of ERBFs in the chromosomal protein fraction extracted with 4 M GdnHCl (CP4) after large scale purification. These protein fractions were further purified by CL-Sepharose 6B column chromatography which resolved fractions from CP4 that recognized the ER bound by estrogen only or antiestrogen only. Thus, these hydrophobic chromosomal proteins enhanced the binding of the ER to reconstituted chromatin. To further investigate the interaction of ERBFs with ER, gel mobility shift assays were performed. The highly purified CP4 fraction with ERBF activity in the binding assay with reconstituted chromatin caused an increase in the formation of the retarded ER-estrogen responsive element (ERE) band. Thus, chromatin contains specific ERBFs for ER bound by estrogen which enhance the binding of ER to genomic DNA and a target ERE sequence. Further purification of the CL-Sepharose fraction with ERBF activity was achieved by preparative SDS-PAGE. ERBF activity was attributed to proteins with approximate molecular weights of 16,000, 13,000, and 12,000 and a pl of > 9.0. Peptides were partially sequenced by Edman degradation and were found to have identity with histones H2B and H4. A 17 kDa protein without ERBF activity was identified as H3. Since these histones were not readily extracted from chromatin with 3 M NaCl or 1-3 M GdnHCl, we postulate that some ERBFs may be histone variants or modified histones that display a very high affinity for DNA and ER.

Interaction of the antiestrogen [3H]H1285 with the two forms of the molybdate-stabilized calf uterine estrogen receptor

The high affinity antiestrogen [3H]H1285 bound to the cytosol calf uterine estrogen receptor dissociated very slowly (t 1/2 approx 30 h at 20 degrees C) and did not demonstrate a change in dissociation rate in the presence of molybdate, which is characteristic of [3H]estradiol-receptor complexes. [3H]H1285-Receptor complexes sediment at approx 6S on 5-20% sucrose density gradients containing 0.3M KCl with or without 10 mM molybdate. This is in contrast to [3H]estradiol-receptor complexes which sedimented at approx 4.5S without molybdate and at approx 6S with molybdate. These results suggest a physicochemical difference in the estrogen receptor when occupied by antiestrogens versus estrogens. We recently reported that the cytoplasmic uterine estrogen receptor, when bound by estradiol and prepared in 10 mM molybdate, eluted from DEAE-Sephadex columns as Peak I (0.21 M KCl) & Peak II (0.25 M KCl). However, [3H]H1285 bound to the estrogen receptor eluted only as one peak at 0.21 M KCl, also suggesting that the initial interaction of antiestrogens with the estrogen receptor is different. We have extended these studies and report that H1285 can compete with [3H]estradiol for binding to both forms of the estrogen receptor and [3H]H1285 can bind to both forms if the unoccupied receptor is first separated by DEAE-Sephadex chromatography. However, if the receptor is first bound by unlabeled H1285, eluted from the column and post-labeled by exchange with [3H]estradiol, only one peak is measured. Thus, it appears that H1285 binding alters the properties of the receptor such that all receptor components seem to elute as one form. These partially purified [3H]H1285-receptor complexes obtained from DEAE-Sephadex columns sedimented as 5.5S in sucrose density gradients in contrast to the sedimentation values for the [3H]estradiol-receptor components eluting as Peak I (4.5S) and Peak II (6.3S). These differences in the physicochemical characteristics of the estrogen receptor when bound by estrogen versus antiestrogens may be related to some of the biological response differences induced by these ligands.

Nuclear Acceptor Sites: Interaction with Estrogen-Versus Antiestrogen-Receptor Complexes

Although it is generally accepted that an interaction between steroid hormone receptors and nuclear acceptor sites is necessary for steroid induction of gene expression, the components of specific nuclear acceptor sites are still being investigated. There have been numerous models for steroid receptor interaction with target cell nuclear components, such as binding to nuclear membranes, RNA, nuclear matrix, DNA sequences and acceptor protein-DNA complexes. None of these models are mutually exclusive of one another.