Switch from antagonist to agonist of the androgen receptor bicalutamide is associated with prostate tumour progression in a new model system.

Advanced prostate cancer is treated by androgen ablation and/or androgen receptor (AR) antagonists. In order to investigate the mechanisms relevant to the development of therapy-resistant tumours, we established a new tumour model which closely resembles the situation in patients who receive androgen ablation therapy. Androgen-sensitive LNCaP cells were kept in androgen-depleted medium for 87 passages. The new LNCaP cell subline established in this manner, LNCaP-abl, displayed a hypersensitive biphasic proliferative response to androgen until passage 75. Maximal proliferation of LNCaP-abl cells was achieved at 0.001 nM of the synthetic androgen methyltrienolone (R1881), whereas 0.01 nM of this compound induced the same effect in parental cells. At later passages (> 75), androgen exerted an inhibitory effect on growth of LNCaP-abl cells. The non-steroidal anti-androgen bicalutamide stimulated proliferation of LNCaP-abl cells. AR protein expression in LNCaP-abl cells increased approximately fourfold. The basal AR transcriptional activity was 30-fold higher in LNCaP-abl than in LNCaP cells. R1881 stimulated reporter gene activity in LNCaP-abl cells even at 0.01 nM, whereas 0.1 nM of R1881 was needed for induction of the same level of reporter gene activity in LNCaP cells. Bicalutamide that acts as a pure antagonist in parental LNCaP cells showed agonistic effects on AR transactivation activity in LNCaP-abl cells and was not able to block the effects of androgen in these cells. The non-steroidal AR blocker hydroxyflutamide exerted stimulatory effects on AR activity in both LNCaP and LNCaP-abl cells; however, the induction of reporter gene activity by hydroxyflutamide was 2.4- to 4-fold higher in the LNCaP-abl subline. The changes in AR activity were associated neither with a new alteration in AR cDNA sequence nor with amplification of the AR gene. Growth of LNCaP-abl xenografts in nude mice was stimulated by bicalutamide and repressed by testosterone. In conclusion, our results show for the first time that the nonsteroidal anti-androgen bicalutamide acquires agonistic properties during long-term androgen ablation. These findings may have repercussions on the natural course of prostate cancer with androgen deprivation and on strategies of therapeutic intervention.

Differential cross-talk of estrogen and growth factor receptors in two human mammary tumor cell lines.

Cultured human mammary MCF7 and T47D tumor cell lines were used to test the interference of the partial antiestrogen 4'-hydroxytamoxifen (4-OH-TAM) and the pure antiestrogen ZM 182780 with growth factor (IGF-I, heregulin) signaling pathways. Growth of both cell lines was stimulated by IGF-I (20 ng/ml) or heregulin (3 nM). ZM 182780 effectively blocked growth factor induced as well as basal proliferation of MCF7 cells while the compound was ineffective in interfering with growth factor mitogenic activity in T47D cells. On both cell lines the IGF-I or heregulin- induced proliferation was enhanced further by 4-OH-TAM. This synergism could be inhibited dose-dependently by ZM 182780. When cells were grown in the presence of estradiol plus growth factors, the antiestrogenic potencies of both compounds and the efficacy of ZM 182780 were unaffected, while the efficacy of 4-OH-TAM was reduced. Our data show cell type specific cross-talk between the receptor for estrogen and that for IGF-I or heregulin, which is different in MCF7 and T47D cells, respectively. In MCF7 cells with demonstrable cross-talk, a clear superiority exists for a pure antiestrogen over a partial agonist in interfering with growth factor mitogenic activity.

Recent developments in molecular action of antihormones.

Antihormones are by definition antagonists of steroid hormone action. They interact with the ligand binding domains of steroid hormone receptors and competitively inhibit the action of the receptors by mechanisms that are not quite understood. In certain cases antihormones also exhibit agonistic activity especially in connection with certain naturally occurring receptor mutants. These observations together with findings of indiscriminate interaction of antihormones with several classes of steroid receptors have necessitated a search of more effective and reliable antihormones. Recent advances in the resolution of the crystal structure of the ligand binding domains of certain members of the steroid receptor family and identification of non-liganded activation of steroid receptors have produced considerable information that can be harnessed into a fruitful search for a new generation of antihormones.

Progesterone receptor repression by estrogens in rat uterine epithelial cells.

Measurements performed using cell lines or animal tissues have shown that the progesterone receptor (PR) can be induced by estrogens. By use of immunohistochemistry we studied the effects of estrogens on the PR levels in the individual cell types of the target organs uterus and breast. In the uteri of rats, ovariectomy induced a decrease in PR immunoreactivity within the myometrium and outer stromal cell layers. In contrast, in the uterine luminal and glandular epithelium and surrounding stromal cell layers the PR immunoreactivity was significantly enhanced. The same picture emerged when intact rats were treated with the pure estrogen receptor antagonist, ZM 182780 (10 mg/kg/d). Treatment of ovariectomized rats with estradiol resulted in high PR levels in the myometrium and stroma cells but low PR immunoreactivity in the epithelial cells. The ER-mediated repression of the PR immunoreactivity was evidently restricted to the uterine epithelium, as we found that in the epithelial cells of the mammary gland and in cells of N-nitrosomethylurea-induced mammary carcinomas the PR expression was induced by estrogens and was blocked by the pure antiestrogen ZM 182780. These results clearly show that in the rat the activated ER induces diverging effects on PR expression in different cell types even within the same organ.

Relation of oestradiol-mediated growth stimulation with the expression of c-erbB-2 protein in xenotransplanted oestradiol-receptor-positive and -negative breast carcinomas.

Attempts were made to correlate growth effects induced by oestradiol and tamoxifen with the hormonal regulation of c-erbB-2 protein in experiments in vivo. We report here the responsiveness of four xenotransplanted oestrogen-receptor(ER)-positive and four ER-negative human mammary carcinomas to oestradiol and tamoxifen. Oestradiol in a dose of 0.5 mg/kg significantly increased the growth of the ER-positive mammary carcinomas 3366, MCF-7, 4134 and 4049, but not the ER-negative tumours 4000, 4296 and MT-3. However, within the group of the ER-negative breast carcinomas the tumour 4151 ES deviates from this growth behaviour, as we could prove an estrogen induced growth. The stimulation of tumour growth by oestradiol was always accompanied by a down-regulation of c-erbB-2 protein both in the ER-positive mammary carcinomas and in the ER-negative mammary carcinoma 4151 ES. Tamoxifen significantly inhibited the growth of the ER/PR-positive mammary carcinomas 3366 and MCF-7 but not the ER-positive/PR-negative mammary carcinomas 4049 and 4134. In the group of ER-negative mammary carcinomas only the growth of the oestrogen-responsive tumour 4151 ES was significantly inhibited by tamoxifen. The inhibition of tumour growth by tamoxifen was correlated with a reversion of the oestradiol-induced down-regulation of c-erbB-2, also in the ER-negative/oestradiol-responsive mammary carcinoma 4151 ES. From our results we hypothesize that the oestrogen-dependent growth of ER-negative breast carcinoma 4151 ES could also be correlated with the oestradiol-regulated expression of c-erbB-2 protein.

The future of antihormone therapy: innovations based on an established principle.

Endocrine therapy of mammary and prostate cancer has been established for decades. The therapies available to block sex-hormone-receptor-mediated tumor growth are based on two principles: (i) ligand depletion, which can be achieved surgically, by use of luteinizing-hormone-releasing hormone analogues or inhibitors of enzymes involved in steroid biosynthesis or by interfering with the feedback mechanisms of sex hormone synthesis at the pituitary/hypothalamic level; (ii) blockade of sex hormone receptor function by use of antihormones. The antiestrogen tamoxifen, which is the compound of choice for the treatment of mammary carcinoma, has the drawback of being a partial agonist. A complete blockade of estrogen receptor (ER) function can be achieved by a new class of compounds, pure antiestrogens. In contrast to aromatase inhibitors, pure antiestrogens are able to block ER activation by ligands other than estradiol and can also interfere with ligand-independent ER activation. In addition to estradiol, progesterone has a strong proliferative effect in mammary carcinomas. Antiprogestins are promising new tools for clinical breast cancer therapy. These compounds clearly need a functionally expressed progesterone receptor to block tumor growth, but there is strong experimental evidence that their tumor inhibition is based on more than just progesterone antagonism. The ability of these compounds to induce tumor cell differentiation that leads to apoptosis is unique among all other endocrine therapeutics. In prostate tumors that have relapsed from current androgen-ablation therapies the androgen receptor (AR) is still expressed and, compared to the primary tumors, its level is often even enhanced. Mutated AR that can be activated by other compounds such as adrenal steroids, estrogens, progestins and even antiandrogens have been detected in recurrent tumors. Thus, relapse of tumors under the selective pressure of common androgen-ablation therapies can be caused by acquired androgen hypersensitivity and AR activation by ligands other than (dihydro-)testosterone. There is a clinical need for future compounds that produce a complete blockade of AR activity even in recurrent tumors. Preclinical experiments indicate that combination therapy as well as the extension of endocrine treatments to several other tumor entities are promising approaches for further developments. Examples are the combination of antiestrogens and antiprogestins for breast cancer treatment, or the treatment of prostate carcinomas with antiprogestins.

Antiprogestins inhibit growth and stimulate differentiation in the normal mammary gland.

Antiprogestins possess a potent antitumor activity in hormone-dependent experimental breast cancer models. Though the underlying mechanism is not clear, induction of functional differentiation seems to be a major event. This study attempts to test directly for antiproliferative and differentiation promoting activities of antiprogestins on the normal mammary gland. To this end, whole organ cultures of mammary glands from estradiol/progesterone-primed virgin mice maintained in a serum-free medium with aldosteron, prolactin, insulin, and hydrocortisone were exposed to the antiprogestin ZK114043. A 4-day treatment of organ cultures led to a strong inhibition of epithelial DNA synthesis. In parallel, ZK114043 caused alveolar cells to acquire a more differentiated phenotype distinguished by secretory active alveoli composed of single cell layers with increased fat droplet accumulation and enhanced expression of the milk proteins beta-casein and whey acidic protein (WAP). Particularly strong effects were found on the expression of mammary-derived growth inhibitor (MDGI). Both half-maximal inhibition of epithelial DNA synthesis and stimulation of MDGI mRNA expression were found at about 5 ng/ml of ZK114043. Presence in the medium of 5 micrograms/ml hydrocortisone rendered antiglucocorticoid effects of ZK114043 highly unlikely. Furthermore, prevention of action of ZK114043 by the progesterone agonist R5020 and ZK114043 stimulated expression of beta-casein and MDGI mRNA in cultured glands of 10-week-old unprimed virgin mice suggest a progesterone receptor-mediated mechanism of antiprogestin action. Two other antiprogestins, Mifepristone and Onapristone, likewise stimulated MDGI expression. The data provide direct evidence that antiprogestins act like a differentiation factor in the normal mammary gland.

Estrogen action on hepatic synthesis of angiotensinogen and IGF-I: direct and indirect estrogen effects.

In the present study effects of estrogens (natural estradiol and synthetic ethinyl estradiol) on liver derived proteins (angiotensinogen, IGF-I) were investigated in vivo in ovariectomized rats and in vitro in a rat hepatoma cell line (Fe33). The aim of this study was to establish both an animal and an in vitro model for quantification of the hepatic activity of given estrogenic compounds, and to study underlying mechanisms as regards the question of direct or indirect mode of estrogen action. In ovariectomized rats subcutaneous (s.c.)-treatment for 11 days with either estradiol (E2) or ethinyl estradiol (EE) (dose range 0.1-3 micrograms/animal/day) induced a comparable dose-dependent increase in uterine weight indicating a similar estrogenic potency of the two estrogens. Equipotency was also found as regards the effects on IGF-I plasma levels which dose-dependently decreased by about 50% at the highest dose tested (3 micrograms/animal/day). The decrease in IGF-I serum levels was accompanied by a significant 40% decrease in liver IGF-I mRNA. In contrast angiotensinogen plasma levels were affected only by EE (60% increase for the 3 micrograms/animal/day dose) but not by E2. When rats, in addition to ovariectomy, were also hypophysectomized (substituted with human growth hormone and dexamethasone) angiotensinogen again increased by 80% upon administration of 3 micrograms/animal/day EE, whereas IGF-I remained unaffected by EE. In a rat hepatoma cell line (Fe33) which is stably transfected with an estrogen receptor expression plasmid, 10 nmol/l EE for 24 h caused a 2.4-fold increase in angiotensinogen mRNA level. We conclude from our studies that estrogen effects on angiotensinogen serum levels in the rat are direct effects via the hepatic estrogen receptor, whereas estrogen effects on IGF-I serum levels are indirect effects, the primary target of estrogen action being probably the pituitary. The changes in angiotensinogen serum levels in the rat model are comparable to the situation in humans indicating the rat model and the Fe33 model to be useful tools to study the hepatic activity of estrogenic compounds.

Gp80 (clusterin; TRPM-2) mRNA level is enhanced in human renal clear cell carcinomas.

The gp80 glycoprotein complex (clusterin, apolipoprotein J, TRPM-2) is a widely expressed protein that has been attributed functions in tissue remodelling, immune defense and transport of lipids and biologically active peptides. The expression of the protein appears to be elevated in several neurodegenerative, apoptotic and malignant processes. We show here that in patients with renal clear cell carcinoma gp80 mRNA is 3-fold overexpressed in tissue of the tumors compared with adjacent non-tumor tissue.

Molecular cloning of gp 80, a glycoprotein complex secreted by kidney cells in vitro and in vivo. A link to the reproductive system and to the complement cascade.

cDNA clones coding for the gp 80 heterodimeric glycoprotein complex secreted constitutively at the apical surface of Madin-Darby canine kidney (MDCK) cells have been isolated from MDCK cDNA libraries in lambda gt11 and lambda gt10. The cloned sequences encode a polypeptide chain of 445 amino acids. The deduced amino acid sequence of the gp 80 protein reveals 80% homology to rat SGP-2, a major secretory protein of the testes epithelium and 83% homology to SP-40,40, a human complement-associated protein. SGP-2 and SP-40,40 have been proposed to be serum and seminal forms of the same protein. The sequence homology as well as the results of Southern and Northern blot analyses and immunological studies suggest that gp 80 is the canine homolog of the rat SGP-2 and the human SP-40,40. The protein is expressed in the embryonic kidney already early during organogenesis. In the adult kidney the protein has been localized along the luminal surfaces of the proximal and distal tubule and the collecting duct cells.

The role of carbohydrates in vectorial exocytosis. The secretion of the gp 80 glycoprotein complex in a ricin-resistant mutant of MDCK cells.

In the polarized epithelial Madin-Darby canine kidney (MDCK) cell line an 80 kDa glycoprotein complex (gp 80) is sorted into the apical pathway of exocytosis and is secreted constitutively at the apical cell surface. The unglycosylated form of the protein complex is secreted in a nonpolar fashion at both surface domains [(1987) J. Cell. Biol. 105, 2735-2743]. Using ricin-resistant MDCK cells the role of the terminal galactose and sialic acid residues in the sorting of the gp 80 complex was analysed. The results suggest that the carbohydrate cores, rather than the ultimate or penultimate sugar residues, play a critical role in the intracellular transport of this protein.

Microtubules are involved in the secretion of proteins at the apical cell surface of the polarized epithelial cell, Madin-Darby canine kidney.

Microtubule-disrupting drugs (nocodazole, colchicine) and cytochalasin D, which inhibits the polymerization of the actin microfilaments, were used to study the role of the cytoskeleton in protein secretion in the polarized Madin-Darby canine kidney (MDCK) epithelial cells. Two proteins were analyzed. The gp 80 glycoprotein complex, which in untreated cells is sorted into the apical pathway and lysozyme, which is released randomly at both cell surfaces in transfected MDCK cells. Our results show that cytochalasin D has no influence on the transport of the gp 80 complex and lysozyme to either cell surface. However, in the presence of nocodazole or colchicine the secretion of both proteins at the apical cell surface is reduced by 50% with a concomitant increase in the basolateral release. These data suggest that microtubules are necessary for an efficient secretion of proteins at the apical cell surface of MDCK cells. In regard to the yet unresolved discrepancy concerning the involvement of microtubules in the transport of membrane proteins to the apical surface of MDCK cells, our results are consistent with the data of Rindler et al. (Rindler, M. J., Ivanov, I. E., and Sabatini, D. D. (1987) J. Cell. Biol. 104, 231-241) who observed a nonpolarized delivery of the influenza virus hemagglutinin in the presence of nocodazole or colchicine.

The influence of pH on the vesicular traffic to the surface of the polarized epithelial cell, MDCK.

The effect of pH on the secretion of the gp 80 glycoprotein complex and lysozyme from MDCK cells was examined by treatment of the cells with either NH4Cl, chloroquine or monensin. In untreated cells gp 80 is sorted with approximately 75% efficiency into the apical pathway. Lysozyme is secreted in a nonpolar fashion at both cell surfaces. Treatment of the cells with the drugs had nearly identical effects on the transport kinetics and on the ratio of the proteins released at the two plasma membrane domains. At increasing drug concentrations, the transport of both proteins to the apical and the basolateral cell surface was equally retarded. Furthermore, we observed a dose-dependent decrease in the amount of gp 80 and lysozyme released at the basolateral cell surface, which was accompanied by a nearly equivalent increase in the secretion of the two proteins at the apical plasma membrane domain. A twofold rise in the apical to basolateral ratio was already found at drug concentrations which only marginally affected the kinetics of transport. These results show that an increase in intravesicular pH not only redirects secretory proteins sorted into the basolateral pathway (Caplan et al. Nature, 329, 632 (1987] but also secretory proteins devoid of sorting information for that pathway, presumably by modulating the vesicular traffic to the basolateral cell surface.

Constitutive apical secretion of an 80-kD sulfated glycoprotein complex in the polarized epithelial Madin-Darby canine kidney cell line.

The biosynthesis, processing, and apical secretion of a group of polypeptides (Kondor-Koch, C., R. Bravo, S. D. Fuller, D. Cutler, and H. Garoff. 1985. Cell. 43:297-306) are studied in MDCK cells using a specific polyclonal antiserum. These polypeptides are synthesized as a precursor protein which has an apparent Mr of 65,000 in its high mannose form. This precursor is converted into a protein with an apparent Mr of 80,000 containing complex carbohydrates and sulfate. After intracellular cleavage of the 80-kD protein, the 35-45-kD subunits are secreted as an 80-kD glycoprotein complex (gp 80) linked together by disulfide bonds. Secretion of the protein complex occurs by a constitutive pathway at the apical surface of the epithelial monolayer. Since the immediate post-translational precursor, the 65-kD protein, is hydrophilic in nature as shown by its partitioning behavior in a phase-separated Triton X-114 solution, gp 80 is segregated into the apical exocytotic pathway as a soluble molecule. The proteolytic maturation of gp 80 is blocked in the presence of chloroquine and its secretion is retarded. The 80-kD precursor is released at the apical cell surface, demonstrating that proteolytic processing is not necessary for the apical secretion of this protein. If N-glycosylation is inhibited by tunicamycin treatment the protein is secreted in equal amounts at both cell surfaces, indicating a role of the carbohydrate moieties in the vectorial transport of this protein.