Search

We have examined the effect of progestin on the regulation of cellular progesterone receptor (PR) levels and have used dense amino acid-density shift experiments to determine the mechanism by which progestin markedly decreases PR. We have utilized T47D human breast cancer cells which contain high levels of PR and are progestin responsive. When these cells are exposed to the progestin R5020, there is a time- and concentration-dependent decrease in PR levels. Experiments with different concentrations of R5020 reveal that the rate and extent of PR decrease reflect the time course of receptor occupancy and the fractional saturation of receptor. With a high concentration of ligand (20 nM) that labels all receptors rapidly, reductions in PR levels (processing) occur immediately and proceed rapidly to levels that are 15-20% of the initial; at lower concentrations (5 nM), where it takes several hours to achieve full saturation of receptors, there is a delay before the maximal rate of processing develops and then continues to achieve final receptor levels that are 15-20% of the initial; with a low concentration of ligand (0.5 nM), binding is even slower and never reaches full receptor saturation, with the consequence that processing is not only delayed but also less complete. Immunochemical detection of PR with a monoclonal antibody (B39) reveals a good correspondence between the loss of immunoreactive and hormone binding PR, and analysis of the A (Mr 85,000) and B (Mr 115,000) receptor forms on Western blots demonstrates that both A and B receptor forms are reduced after exposure to R5020. Density labeling of PR by biosynthetic incorporation of 2H, 13C, 15N (dense) amino acids reveals that PR turns over with a half-life of 21 h in control cells. In cells exposed to 20 nM R5020, PR levels decline and receptor half-life is reduced to 6 h. In addition, there is also a time-dependent decrease in the rate constant of PR synthesis, k8, which decreases to less than 10% of its initial value after 24 h of R5020 exposure. Thus, the R5020-evoked reduction in PR levels in this progestin-sensitive cell line is due both to a marked increase in the rate of receptor degradation as well as a dramatic decrease in the rate of receptor synthesis.

To develop an affinity label for the estrogen receptor that would be an estrogen agonist, rather than antagonist, we prepared several aziridine derivatives of the potent nonsteroidal estrogen hexestrol [3R,4S)-3,4-bis(4-hydroxyphenyl)hexane) bearing an aziridine function on the side chain. Three functional groups link the hexestrol ligand and the aziridine: a carbonyl group (ketone or ester), a thioether, or a methylene chain. The apparent competitive binding affinity of these derivatives for the estrogen receptor ranges from 1.8% to 25% that of estradiol, and most of them bind in a time-dependent, irreversible manner with the receptor, although the rate and efficiency of this binding vary widely, often with relatively small changes in structure. This is consistent with the irreversible attachment requiring a precise alignment of activating and reacting residues in the binding site of the receptor. The estrogenic and antiestrogenic activity of these aziridine derivatives was investigated in MCF-7 human breast cancer cells. Most of the compounds are agonists, with one being an antagonist. The derivative (6R,7S)-1-N-aziridinyl-6,7-bis(4-hydroxyphenyl)-5-nonanone (keto-nonestrol aziridine 3) appears to have the most ideal behavior of the estrogenic affinity labeling agents prepared: It is an agonist, and it binds to receptor irreversibly, efficiently, and quite rapidly.

Previously, we found that Phenol Red, a pH indicator dye commonly used in tissue culture media, had weak estrogenic activity, demonstrable by competitive binding to the estrogen receptor, stimulation of the growth rate of human breast cancer (MCF-7) cells, and elevation of progesterone receptor levels in these cells. We have now examined in more detail the source of this estrogenic activity, present in commercially available preparations of Phenol Red. By high performance liquid chromatography and solvent partitioning, we find that the receptor binding and growth promoting activity does not correspond to the indicator dye itself (phenolsulfonphthalein), but rather to more lipophilic impurities present in these preparations. There are numerous such impurities, many of which show some competitive binding activity, but the major receptor binding activity is accounted for by a single impurity component. Commercial preparations of Phenol Red can be purified by ether extraction of the sodium salt, whereby 95-99% of the lipophilic estrogenic impurities are removed, and the growth stimulating activity towards MCF-7 cells is reduced.

Analysis of newly labeled soluble proteins in uteri of mature rats throughout the estrous cycle indicates that the relative rate of synthesis of the uterine estrogen-induced protein, IP, is high at proestrus, when endogenous estrogen secretion is maximal; it is not synthesized at detectable levels at estrus and metestrus; and some minimal synthesis is seen in diestrus uteri. Injection of exogenous estrogen results in some increase in the IP synthesis rate at proestrus; slight induction of IP synthesis at estrus; and maximal induction (as great as that induced in the mature, ovariectomized uterus by estrogen) at metestrus and diestrus. Studies in the immature (day 21-24) rat, aimed at determining the possible causes of the uterine recalcitrance to exogenous estrogen seen at estrus, indicate that one can reproduce in the immature uterus a period of feeble responsiveness to a second injection of estrogen after exposure to a first, high dose of estrogen. After a single injection of estrogen, there is a lag between the gradual return of full IP inducibility (requiring over 40 hours to reach the control level) and the return of nuclear-translocatable receptor (at control level by 24 hr). This suggests that in addition to the presence of nuclear-translocatable receptor, the response to a second injection of estrogen is dependent upon other factors whose replenishment and/or reactivation is slower than that of the receptor.

In order to understand differences in the mechanism of action of estrogens and antiestrogens, it is essential to make comparisons between compounds with similar duration of action. Hence, in these studies, we compare the action of a long-acting estrogen (17alpha-ethinyl estriol-3-cyclopentyl ether, EE3CPE) and a long-acting antiestrogen (U-11,100AUA) on the immature rat uterus and analyze different dosage regimens (single and multiple injections) in studying the effects of these compounds on the uterine estrogen receptor and on uterine growth and sensitivity to estradiol. During the first 24 h, UA (50 microng) and EE3CPE (5 microng) evoke remarkably similar receptor distribution patterns and uterine wet weight increase; however, pronounced differences are seen with long-term, multiple injection regimens (every 12 or 24 h for 72 h). Such treatment with UA results in maintenance of high nuclear receptor levels and very low cytoplasmic receptor levels (ca. 10% of total), but no further increase in uterine weight, DNA or protein content, or total receptor content beyond 24-48 h. In contrast, multiple injections of EE3CPE produce not only a prolonged nuclear retention of receptor, but a progressive increase in total receptor content in the tissue and 35-50% of total receptor is cytoplasmic; uterine weight and DNA and protein content also continue to increase markedly above the 24 h level, and responsiveness to estradiol is maintained. However, regardless of whether the uterus continues to grow (as the EE3CPE) or stops growing after 24-48 h (as with UA), the receptor content on a cell basis is similar. Hence, uterine responsiveness to estradiol and continued uterine growth appear not to be related to the total content of receptor per cell, but rather are correlated with the cytoplasmic receptor level within the cell. As there is a continuous translocation of cytoplasmic receptor to the necleus in the growing uterus, the antagonistic action of antiestrogens appears to derive from their ability to effect a marked perturbation in the subcellular distribution of receptor, whereby very little of receptor (ca. 10%) is cytoplasmic, and further estrogen receptor accumulation (most likely synthesis) is blocked.

Tert-butylphenoxyethyl diethylamine (BPEA), a compound synthesized by us, was designed to incorporate features important in binding to antiestrogen-binding sites (AEBS) while lacking features important in binding to the estrogen receptor (ER). With this compound, we have addressed the question of the role of AEBS in mediating the growth modulatory effects of antiestrogens. BPEA has an affinity for AEBS 6% that of tamoxifen and an affinity for ER less than 0.0003% that of estradiol. BPEA (10(-11)-10(-6) M) had no effect on the growth of MCF-7 breast cancer cells and no effect on inhibition of the growth of MCF-7 cells by different concentrations of the antiestrogen tamoxifen. In addition, BPEA (even at doses of 1 mg/day X 50 g rat) exhibited no uterotropic or antiuterotropic activity in immature rats and had no influence on the agonistic or antagonistic activity of varying concentrations of tamoxifen on uterine weight. Hence, we conclude that occupancy of AEBS, at least by BPEA, does not modulate growth of the uterus or breast cancer cells and does not influence the potency of tamoxifen as an antiestrogen. These findings raise serious doubts about the role of the AEBS in mediating directly the growth modulatory effects of antiestrogens.

We have prepared two analogs of the antiestrogen tamoxifen that incorporate known DNA-crosslinking functions, a chloroethyl nitrosourea and a nitrosocarbamate moiety, and we have tested their bioactivities in cultures of human breast cancer cells. Both compounds bind to the estrogen receptor from MCF-7 cells, with relative binding affinities of 0.18% for the nitrosocarbamate derivative and 0.35% for the nitrosourea derivative, while the affinity of tamoxifen is 1.8%, and that of estradiol is set at 100%. The tamoxifen-nitrosocarbamate compound demonstrated a dose-related cytotoxicity by the colony formation and cell proliferation assays that was not blocked by estradiol in either estrogen receptor-positive MCF-7 cells or estrogen receptor-negative MDA-MB-231 cells, and thus, was not studied further. Tamoxifen-nitrosourea (TAM-NU) showed dose-related cytotoxicity in MCF-7 cells that was blocked by estradiol, whereas its activity in MDA-MB-231 cells was unaffected by estradiol. N-2-(4-t-butylphenoxy)ethyl-N'-chloroethyl-N'-nitrosourea (BPE-NU), a control compound which contains the nitrosourea moiety but does not bind to the estrogen receptor, had no effect on cell proliferation or colony formation in MCF-7 cells, but was very inhibitory in the receptor-negative MDA-MB-231 cells. In contrast, TAM-NU was more active in the receptor-positive MCF-7 cells than in the MDA-MB-231 line. Thus, because TAM-NU appears to be active selectively against the receptor-positive cell line, and because this activity is suppressible by estradiol, its cytotoxic effect seems to be mediated via the estrogen receptor. However, since TAM-NU is active only in prolonged treatment protocols, it appears likely that its cytotoxic activity results from the hormone antagonistic effect of the hydrolysis product of TAM-NU (bis-desmethyltamoxifen), rather than from a direct receptor-mediated, DNA-directed cytotoxic action of TAM-NU itself. This study stresses the need for the use of appropriate control compounds and cell systems in order to assess whether the toxic activity displayed by hormone-cytotoxic conjugates is mediated by receptor interactions and whether it operates through the intended toxic mechanism.

We have examined the effects of reversibly and irreversibly binding estrogenic and antiestrogenic ligands for the estrogen receptor on pS2 RNA accumulation in MCF-7 human breast cancer cells and on pS2-chloramphenicol acetyl transferase (CAT) fusion gene expression in transfected MCF-7 cells. In MCF-7 cells grown in the absence of estrogens, the reversibly binding estrogen, estradiol, and the affinity labeling estrogen, ketononestrol aziridine, KNA, evoked a 13-fold increase in pS2 RNA level. The reversibly binding antiestrogen trans-hydroxytamoxifen and the affinity labeling antiestrogens tamoxifen aziridine or desmethylnafoxidine aziridine behaved as partial agonists/antagonists. In thymidine kinase-chloramphenicol acetyltransferase (tk-CAT) fusion genes containing a 1000 base pair fragment of the pS2 5'-flanking region encompassing the estrogen responsive element of the gene [pS2 (-1100/-90) tk-CAT], estradiol and ketononestrol aziridine evoked a marked stimulation of CAT activity and, in transfected cells grown in both the presence or absence of the weak estrogen phenol red, the antiestrogens behaved as partial agonists/antagonists. This pS2 5'-flanking region displayed both estrogen-dependent and estrogen-independent enhancer activity as monitored by stimulation of CAT activity. Hormonal regulation of the transfected pS2 fusion gene was similar to that observed in the native pS2 gene of MCF-7 cells; however, antiestrogens, while still partial agonists-antagonists, were relatively more agonistic on the transfected fusion gene than on the native gene. One antiestrogen (ICI 164,384) that behaved as a pure estrogen antagonist on the native gene was a partial agonist-antagonist of pS2 gene expression in the plasmid. This study illustrates that the hormonal regulation of the pS2 gene, as characterized by the agonist-antagonist balance of estrogens and antiestrogens, is influenced by the DNA context of the pS2 estrogen responsive element. Also, the fact that estrogens and antiestrogens that form covalent bonds with the estrogen receptor modulate activity of the native pS2 gene and the pS2-tk-CAT fusion gene in a manner similar to that of their reversibly binding counterparts suggests that it may be possible to use these irreversibly binding ligands to follow the interaction of hormone-receptor complexes with regions regulating estrogenic stimulation of the pS2 gene.

Tamoxifen is used widely in the treatment of endocrine-responsive breast cancers in humans. Studies were undertaken to examine the biological character (estrogenic-antiestrogenic properties) and estrogen receptor (ER) interaction of the cis- and trans-isomers of tamoxifen and hydroxytamoxifen in MCF-7 human breast cancer cells. For each compound, the following parameters were monitored: affinity for ER and effects on cellular ER levels; stimulation-inhibition of cell growth, plasminogen activator activity, and cellular progesterone receptor levels; and isomer interconversion and metabolism in vitro. The relative binding affinities of the compounds cis-tamoxifen, trans-tamoxifen, cis-hydroxytamoxifen, and trans-hydroxytamoxifen for cytosol ER were 0.3, 2.5, 1.8, and 310%, respectively, in which the affinity of estradiol is considered 100%. cis-Tamoxifen behaved as a weak estrogen agonist in all assays, while trans-tamoxifen was an effective estrogen antagonist. cis-Tamoxifen behaved like estradiol in stimulating MCF-7 cell growth and increasing plasminogen activator activity and cellular progesterone receptor content, although very much higher concentrations of cis-tamoxifen (10(-6) M) were needed to achieve the levels of stimulation observed with 10(-10) M estradiol. trans-Tamoxifen and trans-hydroxytamoxifen suppressed cell growth, inhibited plasminogen activator activity of control cells, and suppressed estradiol-stimulation of plasminogen activator activity, and they evoked minimal increases in cellular progesterone receptor levels. trans-Hydroxytamoxifen had a 100-fold increased affinity for ER and was approximately 100-times more potent than was trans-tamoxifen in suppressing cell growth and plasminogen activator activity. cis-Hydroxytamoxifen behaved as an estrogen antagonist, suppressing cell growth and plasminogen activator activity, and it elicited submaximal increases in progesterone receptor levels. This apparently paradoxical behavior of cis-hydroxytamoxifen was shown to be due to the fact that the cis- and trans-hydroxytamoxifens readily undergo isomeric interconversion upon exposure to our cell culture conditions, resulting in substantial accumulation of the higher-affinity trans-hydroxytamoxifen in the nuclear ER fraction of cells. In contrast to the facile interconversion of the hydroxytamoxifen isomers, there is no metabolism or interconversion of the parent compounds cis- and trans-tamoxifen in vitro. Hence, by the criteria we have used, the biological characters of trans-tamoxifen and trans-hydroxytamoxifen are similar, the major difference being the approximately 100-fold enhanced potency of the hydroxylated form. In contrast, cis-t

The human breast cancer cell line MCF-7 responds to estrogens with increased progesterone receptor (PR) levels. In this study, we use dense amino acid density shift analyses to address directly the question of whether estrogen increases PR levels in MCF-7 cells by altering rates of receptor synthesis and/or degradation. Using different concentrations of estradiol (E2), which achieve PR levels that are half-maximal (3 X 10(-11) M F2) or maximal (6 X 10(-11) M E2), we have done sucrose gradient density shift analyses using dense (15N, 13C, 2H) amino acid incorporation to study rates of PR synthesis and degradation. These studies reveal a nonlinear loss of preexisting normal density receptor with time. From kinetic modeling analyses, equivalent rates of degradation are estimated for PR whether maximal or half-maximal levels are maintained, indicating that the major effect of E2 on PR content is to increase the rate of PR synthesis while leaving the degradation rate unaltered. The E2-stimulated increase in PR protein is also associated with increased levels of PR mRNA, as demonstrated by the use of a human PR cDNA probe. These density shift data provide evidence that the increased PR levels after estrogen exposure in MCF-7 cells are the result of an increased rate of receptor synthesis, rather than modulation of the rate of receptor degradation.

High concentrations of androgens in vitro [10(-6) and 10(-7)M 5alpha-dihydrotestosterone (DHT) and testosterone (T)] translocate the estrogen receptor from cytoplasmic to the nuclear fraction of the immature rat uterus, and the androgen translocated sites are capable of eliciting the synthesis of the specific uterine "induced protein" (IP), formerly attributed to estrogenic compounds only. The magnitude of receptor translocation and IP synthesis induction is related to the concentration of androgen, and, at equal concentrations, DHT is more effective than T. Competitive protein-binding asssays with cell-free uterine cytosol indicate that DHT and T bind with barely detectable affinity to the cytosol estrogen receptor detectable affinity to the cytosol estrogen receptor [relative binding ability ca. 0.001% that of estradiol (E2)], and radioactive E2 uptake into whole uterus after pretreatment with androgens indicates that the androgen-translocated nuclear sites are readily filled by E2. DHT and T translocate the estrogen receptor to the nucleus in vitro, and the salt-extracted nuclear receptor is present as a 5 S ("transformed") receptor; however, concentrations of DHT and T that effect translocation are unable to elicit the heat-activated transformation of the estrogen receptor in cell free cytosol under conditions in which low concentrations of E2 fully transform the receptor. Under in vivo conditions, high levels of androgens (5--2,000 mug DHT or T) do not evoke any detectable translocation of the estrogen receptor and do not elicit any IP synthesis induction although uterine weight is increased (greater than 200 mug androgen) in long-term (2--3 day) assays. The analysis of androgen uptake and metabolism indicates that 500 mug or higher doses of DHT or T in vivo result in uterine concentrations of unmetabolized DHT or T equal to those seen after exposure to 5 X 10(-7) - 1 X 10(-6)M DHT or T in vitro. Hence, under conditions where in vivo and in vitro tissue uptake of androgen is equivalent, in vivo androgens are unable to affect the estrogen receptor system as is seen in vitro. These studies indicate that the in vitro and in vivo effects of androgens on the uterus are clearly different and suggest that the actions of androgens on the uterus in vivo are probably not directly mediated through the estrogen receptor system.

We have characterized the molecular properties of the plasminogen activators in different cell types comprising the immature and the estrogen-stimulated rat uterus and in rat uterine luminal fluid. There were two plasminogen activators in the immature (day 20) rat uterus with apparent molecular weights, determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, of 70,000 and 46,000. Both plasminogen activators were present in epithelial and in stromal plus myometrial cell fractions of the immature uterus, and after stimulation by 17 beta-estradiol, no new plasminogen activators were detected in either cell fraction. The Michaelis constants (Km) for the activation of dog plasminogen by extracts from epithelial cells and from stromal plus myometrial cells obtained from either immature or 17 beta-estradiol-stimulated uteri were similar (approximately 11 microM). The maximal velocity (Vmax), normalized to protein concentration, increased 2.5-fold in the stromal plus myometrial cell fraction and 6.5-fold in the epithelial cell fraction, upon hormone stimulation (2 micrograms 17 beta-estradiol/day X rat for 3 days). The greatest concentration of plasminogen activator activity was found in the luminal fluid from estrogen-stimulated uteri, where the Vmax per mg protein was more than 10-fold greater than that in the cell fractions from estrogen-stimulated uteri. The plasminogen activator activity of luminal fluid was inhibited by diisopropyl fluorophosphate and rho-nitrophenyl rho-guanidinobenzoate, was not inhibited by human alpha-1-proteinase inhibitor and human antithrombin III, and was inhibited by high, but not low, concentrations of soybean trypsin inhibitor and bovine pancreatic trypsin inhibitor. These studies indicate that the plasminogen activators in different cell types comprising the uterus are similar and show that the estrogen enhancement of uterine plasminogen activator activity is the result of an increase in Vmax. The presence, upon hormone stimulation, of an apparent concentration gradient of increasing plasminogen activator activity through the uterus from myometrium to epithelium to luminal fluid may be a reflection of the dynamic role of this protease in the physiology of the uterus.

We have examined the effect of short-term and long-term growth in the absence of estrogens on the proliferation rate and estrogen and antiestrogen responsiveness of MCF-7 human breast cancer cells. The removal of phenol red, the pH indicator in tissue culture medium that is weakly estrogenic (Y. Berthois et al., Proc. Natl. Acad. Sci. USA, 83:2496-2500, 1986), immediately slows the cell proliferation rate, and MCF-7 cells grown in phenol red-free medium with charcoal dextran-treated serum for periods up to 1 mo maintain this reduced rate of cell proliferation. In these short-term phenol red-withdrawn cells, estradiol stimulates proliferation markedly and reproducibly, and antiestrogens inhibit estrogen-stimulated proliferation. Antiestrogens by themselves appear as partial agonists/antagonists; at low concentrations they stimulate proliferation weakly, but they show no stimulation at the high concentrations where they fully inhibit estrogen-stimulated proliferation. In contrast to the short-term phenol red-withdrawn cells, cells maintained for several months (5 to 6 mo) in the apparently complete absence of estrogens (no phenol red, with charcoal dextran-treated calf serum) show a markedly increased basal rate of cell proliferation; estradiol is unable to increase this rate of proliferation further, but antiestrogens are able to decrease proliferation. This change in growth pattern is associated with a 3-fold increase in cellular estrogen receptor levels. Despite their differing basal growth rates, cells grown in either the short-term (less than 1 mo) or long-term (greater than 6 mo) absence of estrogens both have progesterone receptor levels that are very low and, in both cases, estradiol increases progesterone receptor levels markedly. Thus, under long-term estrogen-free conditions, there is a dissociation between the stimulation of cell proliferation and of specific protein synthesis (progesterone receptor) by estrogen. The increase in the cell proliferation rate observed in cells grown in the long-term absence of estrogen may reflect altered regulation of growth factor production or altered sensitivity to growth factors in the medium or produced by the cells themselves. Hence, these breast cancer cells adapt significantly to long-term growth in estrogen-free conditions, an observation that may be relevant to understanding the growth of hormone-responsive human breast cancers in vivo.

We have examined the ability of various antiestrogens (AE's) to compete with 3H-spiroperidol for binding to membrane preparations from striatal tissue and anterior pituitary glands of immature female rats in order to determine the affinity of binding of AE's to D-2 dopamine receptors. Scatchard analyses revealed the presence of a single class of high affinity receptor sites in both the striatum and pituitary with a dissociation constant (Kd) of 0.33 nM and 0.40 nM, respectively, for the dopamine antagonist spiroperidol. The AE's tamoxifen, 4-hydroxy-tamoxifen (TAM-OH), CI-628, LY 117018, and a structurally related compound t-butyl-phenoxyethyl diethylamine (BPEA) were all able to compete with spiroperidol for binding to D-2 receptors and demonstrated relative binding affinities of 0.4-0.06%, with spiroperidol set at 100%. Dopamine displayed a lower affinity, 0.01%. Estradiol failed to compete with spiroperidol for D-2 receptor binding while the non-steroidal estrogen diethylstilbestrol (DES) showed very weak competition. For the lipophilic AE's, alteration of the level of their non-specific binding greatly affected their relative affinities in these competitive binding assays. The amine side chain on an aromatic ring appears to be a critical structural requirement in allowing the AE's to bind to the dopamine receptor. The relatively low affinity of AE's for the dopamine receptor and the high degree of interaction of AE's with other proteins suggest that only limited occupancy of D-2 receptors by AE's is likely in vivo.

Although much attention has been paid to the removal of hormones from sera and to the development of serum-free media for studies on hormone-responsive cells in culture, little consideration has been given to the possibility that the media components themselves may have hormonal activity. We have found that phenol red, which bears a structural resemblance to some nonsteroidal estrogens and which is used ubiquitously as a pH indicator in tissue culture media, has significant estrogenic activity at the concentrations (15-45 microM) at which it is found in tissue culture media. Phenol red binds to the estrogen receptor of MCF-7 human breast cancer cells with an affinity 0.001% that of estradiol (Kd = 2 X 10(-5) M). It stimulates the proliferation of estrogen receptor-positive MCF-7 breast cancer cells in a dose-dependent manner but has no effect on the growth of estrogen receptor-negative MDA-MB-231 breast cancer cells. At the concentrations present in tissue culture media, phenol red causes partial estrogenic stimulation, increasing cell number to 200% and progesterone receptor content to 300% of that found for cells grown in phenol red-free media, thereby reducing the degree to which exogenous estrogen is able to stimulate responses. The antiestrogens tamoxifen and hydroxytamoxifen inhibit cell proliferation below the control level only when cells are grown in the presence of phenol red; in the absence of phenol red, the antiestrogens do not suppress growth. The estrogenic activity of phenol red should be considered in any studies that utilize estrogen-responsive cells in culture.

We have used limited proteolysis of affinity-labeled estrogen receptors (ER), coupled with antireceptor antibody immunoreactivity, to assess structural features of ER and the relatedness of ER from MCF-7 human breast cancer and rat uterine cells. MCF-7 ER preparations covalently labeled with [3H]tamoxifen aziridine [( 3H]TAZ) were treated with trypsin (T), alpha-chymotrypsin (C), or Staphylococcus aureus V8 protease prior to electrophoresis on sodium dodecyl sulfate gels. Fluorography revealed a distinctive ladder of ER fragments containing TAZ for each protease generated from the Mr 66,000 ER: for T, fragments of 50K, 38K, 36K, 31K, 29K, and 28K that with longer exposure generated a 6K fragment; for C, fragments of 50K, 38K, 35K, 33K, 31K, 19K, and 18K that with longer exposure generated 14K and 6K fragments; and for V8, ca. 10 fragments between 62K and 28K. Two-dimensional gels revealed charge heterogeneity (two to three spots between pI 5.5 and 6.2) of the 66K ER and the T-generated 28K meroreceptor form. Immunoblot detection with the primate-specific antibody D75P3 gamma revealed that all immunoreactive fragments corresponded to TAZ-labeled fragments but that some small TAZ-labeled fragments (V8-generated forms less than 47K and T-generated forms less than 31K) were no longer immunoreactive. In contrast, use of the antibody H222Sp gamma revealed a correspondence between TAZ-labeled and immunoreactive fragments down to the smallest fragments generated, ca. 6K for T and C and 28K for V8. MCF-7 nuclear and cytosol ER showed very similar digest patterns, and there was a remarkable similarity in the TAZ-labeled and H222-immunoreactive fragments generated by proteolysis of both MCF-7 and rat uterine ER. These findings reveal great structural similarities between the human (breast cancer) and rat (uterine) ER and between nuclear and cytosol ER, indicate charge heterogeneity of ER, and allow a comparison of the immunoreactive and hormone attachment site domains of the ER. The observation that T and C generate a ca. 6K TAZ-labeled fragment that is also detectable with the H222 antibody should be of interest in studies determining the hormone binding domain of the ER and in amino acid sequencing of this region.

The rate at which the estrogen receptor (ER) is synthesized and degraded (turns over) in target cells is a major factor regulating the levels of ER and cell sensitivity to estrogen. We have used the ability of tamoxifen aziridine [(TAZ)(Z)-1-[4-(2-[N-aziridinyl]ethoxy)phenyl]1,2-diphenyl-1-buten e] to affinity label the ER in intact cells to study the dynamics of ER turnover in uterine cells in vitro and in intact rat uteri in vivo. In primary cultures of rat uterine cells (from 21-day-old rats), ER is degraded with a half-life of 3-4 h, as determined by pulse-chase experiments in which ER in cells are covalently labeled with [3H]TAZ (20 nM) and then exposed to a chase of 10(-6) M estradiol. Density shift experiments, in which uterine cells are exposed for different time periods to medium containing dense (15N,13C,2H) amino acids and the shift of receptor from a normal density to a more dense species is analyzed on sucrose gradients, also confirm this rapid turnover for the uterine ER. A similar half-life is obtained for ER in rat uteri in vivo, after intralumenal installation of [3H]TAZ to label effectively ER covalently. Analysis of nuclear ER on sodium dodecyl sulfate-polyacrylamide gels during the chase period reveals the loss of the 65,000 mol wt [3H]TAZ-labeled receptor species; no lower mol wt [3H]TAZ-labeled fragments are observed. Likewise, immunoblot analyses of receptor with the ER monoclonal antibody H222Sp gamma reveal only the 65,000 mol wt receptor species, with no evidence of any smaller receptor forms in cells. Interestingly, incubation of uterine cells with 10(-5) M cycloheximide (which inhibits 98% of protein synthesis) completely arrests ER turnover, suggesting that ER turnover, which is rapid, is probably dependent upon the synthesis of other rapidly turning over proteins. Hence, we conclude that ER turnover is rapid in uterine cells, that turnover of uterine ER occurs at a rate similar to that previously reported for ER turnover in MCF-7 human breast cancer cells, and that ER turnover is similar in uterine cells under in vitro and in vivo conditions. These findings indicate that the receptor is a dynamic and rapidly turning over protein whose level could be adjusted quickly by alterations in the rate of synthesis and/or degradation.

The Upjohn antiestrogen (+/-)-cis-3-[p-(1,2,3,4-tetrahydro-6-methoxy-2-phenyl-1-naphthyl)phenoxy]-1,2-propanediol (2b, U 23469) has been prepared in tritium-labeled form by reduction of an unsaturated dihydronaphthalene precursor with carrier-free tritium gas over a palladium catalyst followed by alkylation with 3-iodo-1,2-propanediol. After extensive chromatographic purification, the final material was obtained with a specific activity of 13 Ci/mmol and a radiochemical purity of 94%. In vivo studies with immature rats show that [3H]2b is slowly converted to a more polar metabolite that is selectively accumulated in the nuclear fraction of the uterus where it is bound to the estrogen receptor. Chromatographic comparisons indicate that this metabolite is the demethylated analogue 2c, a compound that has an affinity for estrogen receptor more than 300 times greater than that of 2b. These studies suggest that the demethylated analogue 2c may be a biologically important metabolite of 2b that is involved in the action of this antiestrogen.