Comparison between the interaction of steroids with [35S]TBPS binding to cerebral cortical and to pituitary membranes: correlation with inhibition of prolactin release.

It has been shown previously that 5 alpha-pregnan-3 alpha-ol-20-one (5 alpha 3 alpha P) can inhibit prolactin release from anterior pituitary gland cells in culture through an interaction with a specific modulatory site on the GABAA receptor complex in anterior pituitary gland membranes. In the present work, this receptor site has been labelled with [35S]t-butylbicyclophosphorothionate ([35S]TBPS) to enable a study of the relative binding affinities (RBA) of different steroids for the GABAA receptor complex to be made. We have found a high correlation (r = +0.88) between the inhibition of [35S]TBPS binding to anterior pituitary membranes and the inhibition of [35S]TBPS binding to cerebral cortical membranes by nine different steroids. There was also a high correlation between the inhibition of prolactin release from anterior pituitary gland cells in culture by these steroids and the inhibition of [35S]TBPS binding to anterior pituitary membranes (r = +0.99) or to cortical membranes (r = +0.81). These observations suggest that the measurement of prolactin release from anterior pituitary gland cells in culture is a good indicator of the functional activity of drugs that bind to the allosteric modulatory TBPS-binding site on the GABAA-receptor complex.

Pharmacology of antiandrogens and value of combining androgen suppression with antiandrogen therapy.

Antiandrogens are compounds able to block the effect of androgens directly on their target cells by inhibiting their binding to the androgen receptor (AR). Two chemical classes of antiandrogens are presently on the market or in clinical trials: steroids (cyproterone, megestrol acetates), and nonsteroids (flutamide, nilutamide). Steroid antiandrogens interact not only with AR but also with progestin and glucocorticoid receptors and thus give rise to progestin and glucocorticoid effects. By contrast, nonsteroid antiandrogens interact only with AR and are thus devoid of other hormonal or antihormonal activities. Nilutamide does not need to be transformed into an active metabolite, unlike flutamide, and interacts with dog, rat, and human prostate AR in vitro. Its kinetics lead to a prolonged interaction with AR in vivo after administration to rats. In prostate cancer treatment, it is necessary to combine an antiandrogen to surgical or chemical (estrogens, LH-RH agonists) castration to obtain a complete suppression of androgens. The antiandrogen will block specifically, at the target site, the trophic effect of adrenal androgens left intact by castration, and the secretion of which can only be suppressed by treatments (adrenalectomy, aminoglutethimide, ketoconazole) that also suppress corticoid synthesis. We have shown that nilutamide counteracts the trophic effect, on the prostate of castrated rats, of adrenal androgens administered continuously (minipumps) at circulating levels similar to those recorded in castrated men. Nilutamide will also impede the flare-up effect of the testosterone increase induced by LH-RH agonists at the beginning of treatment. We have shown in the rat treated with buserelin that the increase in prostate weight observed during the initial days of treatment by the LH-RH agonist can be inhibited by a combined treatment with nilutamide. This combined treatment "nilutamide plus castration" has been tested in an experimental androgen-dependent cancer model, the Shionogi tumor. The administration of nilutamide to mice, castrated twenty-four hours before the inoculation of tumor cells, delayed the appearance of tumors and reduced their number. Finally, the absence of androgen effect and the antiandrogen activity of the product were also demonstrated in human tumor cells in culture (T-47 D cells) transfected with the MMTV androgen-dependent promoter coupled with the CAT reporter gene.

A progesterone metabolite enhances the activity of the GABAA receptor complex at the pituitary level.

The interaction of 5 alpha-pregnane-3 alpha-ol-20-one (5 alpha 3 alpha P), a progesterone metabolite, with the GABAA receptor chloride channel complex was investigated at the pituitary level. In nanomolar concentrations this steroid potentiated the inhibitory effect of muscimol (a GABAA agonist) on prolactin release from rat pituitary cells in culture. In micromolar concentrations 5 alpha 3 alpha P had a direct inhibitory effect, similar to that of muscimol, with an IC50 value of 370 nM. This effect was antagonized by bicuculline, a GABAA antagonist, and by picrotoxin, a chloride ion channel blocker. Its reduced isomer, 5 alpha 3 beta P, and progesterone (Pg) were devoid of activity. Using [35S]t-butylbicyclophosphorothionate ([35S]TBPS) as a ligand, we demonstrated, for the first time, specific barbiturate sites on pituitary membranes, similar to those of the central nervous system, with a Kd value of 25 nM and a Bmax value of 62 fmol/mg protein. 5 alpha 3 alpha P inhibited the binding of [35S]TBPS. In contrast, its 3 beta isomer was inactive. These data show that 5 alpha 3 alpha P enhanced the activity of the GABAA receptor complex at the pituitary level and suggest that its inhibitory effect on prolactin release might be mediated by the barbiturate site or by a closely related site.

New analogues of mifepristone with more dissociated antiprogesterone activities.

Mifepristone (RU 486 or RU 38486) possesses strong antiprogesterone and antiglucocorticoid along with moderate antiandrogen properties, which would limit its use in some therapeutic applications. In a search for more dissociated derivatives, the hydroxy substituent and the propynyl group in position 17 of the RU 486 series was replaced by a spiroether group, which is known to induce specific affinity for the progestin receptor in steroid series. The substituents in the para position of the 11 beta-phenyl group, leading to the most potent derivatives in the RU 486 series, were retained. The new derivatives have been studied in vitro for their relative binding affinities (RBAs) for the steroid receptor and in vivo for their hormonal and antihormonal activities. The selected compounds, RU 46556 and RU 49295 display the following properties: in vitro, like RU 486, they show a strong RBA for the rabbit progestin receptor, but a much lower one for the rat thymus glucocorticoid receptor; in vivo they are about three times more active than RU 486 for inducing abortion in rats, but unlike the latter they are devoid of any antiglucocorticoid activity on the thymus weight in rats. These antiprogesterone effects have been confirmed on the deciduoma formation in rats and on the endometrial proliferation in rabbits. However, in contrast to RU 486 in the latter test, some progestomimetic activity has been observed. RU 46556 and RU 49295 are now under extensive pharmacological study.

How the study of the biological activities of antiandrogens can be oriented towards the clinic.

Antiandrogens can be used in various androgen-dependent diseases. Depending upon the therapeutic indication, they can be administered systemically or topically. Systemic treatment with an antiandrogen will inhibit androgen action not only in the desired target site but also in all other target tissues; thus, it will block the androgen-dependent feedback regulating the secretion (hypothalamo-pituitary-testis axis) or the action (protein factors) of androgens. In contrast, topical treatment (acting through cutaneous receptors or local metabolism) should not produce systemic side effects especially in man. Pharmacological assays which can select antiandrogens irrespective of the mechanism measure changes in the final androgenic response, but they consume a great deal of time and test compound and bear little relation to therapeutic activity. Therefore, the biological strategy that we report here and which, at Roussel-Uclaf, has led to the selection of a systemic and a topical antiandrogen (RU 23908 and RU 38882) has consisted in successively performing: (1) in vitro assays which measure an effect at a specific level in the mechanism of antiandrogen action, e.g. interaction with the androgen receptor. Assessing interactions with other classes of steroid hormone receptor can be used to predict possible hormonal side-effects, (2) in vitro determinations of agonist or antagonist activity, e.g. in pituitary cells (LH response to LHRH) or mammary tumor cells (induction of androgen-dependent proteins), (3) in vivo antiandrogen assays after a single treatment (induction of mouse kidney proteins, rat prostatic binding protein) or after repeated treatment (inhibition of the growth of rat accessory glands or of hamster sebaceous glands), to determine the active dose of the compound and possibly the absence of systemic effects by the topical route, (4) assays in animal models designed to mimic a therapeutic context e.g. for prostate cancer: inhibition of the "flare-up" effect of LHRH-A or of the trophic effect of perfused adrenal androgens on rat prostate, antitumoral activity in experimental cancer models. For hyperseborrhoea and acne: histological and stereological analysis of rat skin biopsies to measure the volume density of the smooth endoplasmic reticulum vesicles of the differentiating cells of the sebaceous gland.

Autoradiographic localization of glucocorticosteroid binding sites in rat brain after in vivo injection of [3H]RU 28362.

The autoradiographic distribution of glucocorticosteroid binding sites in the brain of adrenalectomized rats was studied following in vivo injection of a potent synthetic glucocorticosteroid agonist [3H]RU 28362. Analysis of the autoradiograms revealed a specific and dense labeling in the pyramidal cell layer of the Ammon's horn and in the granular cell layer of the dentate gyrus of the hippocampus. In the hypothalamus, the labeling was particularly high in the paraventricular nucleus (site of CRF synthesis), the arcuate, periventricular and the supraoptic nuclei as well as in the median eminence. Autoradiograms also revealed the presence of [3H]RU 28362 binding sites in several brain regions including the amygdala, the pineal gland, the entorhinal cortex, the interpeduncular, interfascicular and dorsal raphe nuclei, the central grey and the substantia nigra suggesting possible effects of glucocorticosteroids in these structures.

The antiandrogen anandron potentiates the castrating effect of the LH-RH agonist buserelin in the rat.

When buserelin (0.04-25 micrograms/kg/day), a potent LH-RH agonist, was administered s.c. daily for 15 days to male rats, prostate weight decreased after a transient increase in the first days of treatment but never as much as after orchidectomy, since testosterone secretion was never totally suppressed. The combination of the pure nonsteroid antiandrogen Anandron (20 mg/kg/day) with even low doses of buserelin led to an immediate decrease in prostate weight that was complete at 15 days. This could be explained (a) by an additivity of effects: inhibition by Anandron of the action of residual testosterone unsuppressed by buserelin on the prostate and prevention by buserelin of the rebound testosterone increase induced by Anandron: (b) by a potentiation by Anadron of the direct castrating effect of buserelin on the testis since testis weight and testosterone secretion were lowered to a greater extent by the combination than by buserelin alone. Anandron might increase the sensitivity of LH-RH receptors to LH-RH agonist in the testes, as has been shown in the pituitary. In contrast, the combination of the antiandrogenic steroid progestin, cyproterone acetate, with buserelin never potentiated the castrating effect of buserelin on testis weight and testosterone and only partially potentiated prostate atrophy. If such actions also exist in the human, the addition of Anandron to LH-RH agonist treatment would not only counter flare-up and adrenal androgen effects on the prostate but would also lower the doses of peptide or the time required for chemical castration.

Binding and action of glucocorticoids and mineralocorticoids in rabbit mammary gland. Exclusive participation of glucocorticoid type II receptors for stimulation of casein synthesis.

In order to ascertain whether the effect of corticoids upon casein synthesis in pregnant rabbit mammary gland culture is due to interactions with classical glucocorticoid or type I (mineralocorticoid) receptors we have demonstrated the existence of both types of receptors in the tissue and have studied the effects of aldosterone and the specific glucocorticoid agonist RU 28362 upon casein synthesis in tissue culture. Both compounds significantly stimulated prolactin-induced casein synthesis. On dose-response studies RU 28362 proved to be as active as dexamethasone, cortisol was active at intermediate concentrations and aldosterone was the least active. The three glucocorticoids were able to stimulate DNA synthesis in the tissue, but aldosterone had no effect. Finally, RU 486, a potent glucocorticoid antagonist, blocked the action of aldosterone and the other corticoids upon casein synthesis, whereas spironolactone, a mineralocorticoid antagonist, was unable to do so. These results demonstrate that the stimulatory effect of corticoids upon casein synthesis in pregnant rabbit mammary tissue culture is mediated through classical (type II) glucocorticoid receptors. Transferrin accumulation in the tissue was not modified by any treatment, indicating that the action of the steroids was specific for casein, and not a general stimulation of protein synthesis.

Pharmacological and clinical studies of the antiandrogen Anandron.

This paper summarizes the animal and human studies with Anandron available at the time of the meeting. The following was demonstrated in the rat and confirmed in man: interaction of Anandron with the prostatic androgen receptor, antiandrogen activity against testosterone (in particular against the early transient rise induced by LHRH analogs) and adrenal androgens. Thus, as shown in 4 different double blind studies performed in stage D2 prostrate cancer patients, the combination of Anandron with surgical or chemical castration enhanced the beneficial effects of castration alone and thus seems a step forward in the hormonal treatment of prostatic carcinoma.

Design of antiandrogens and their mechanisms of action: a case study (anandron).

The design of a new drug is conditioned by knowledge of the biochemical mechanisms involved in the etiology of the disease to be treated. With regard to endocrine pathologies, such knowledge can be obtained in the clinic from systematic assays of urinary and plasma hormones, enzyme activities and target tissue receptor concentrations. The present paper describes the results of our assays of plasma 3 alpha-androstanediol glucuronide, 5 alpha-reductase and androgen receptor in prostate cancer patients. The activity of the nonsteroid antiandrogen anandron is discussed in relation to these parameters: anandron may inhibit slightly adrenal androgen biosynthesis but, in particular, counters the action of these adrenal androgens on the prostate. It does not inhibit rat prostate 5 alpha-reductase activity but interacts with androgen receptor to exert an antiandrogen action.

Corticosteroid receptors in rat hippocampal sections: effect of adrenalectomy and corticosterone replacement.

Rat brain sections, located at the hippocampal level, were used to study the effect of bilateral adrenalectomy, with or without corticosterone treatment, on the number and affinity of corticosteroid binding sites. Adrenalectomy induces an increase of corticosterone receptor binding sites whereas adrenalectomy followed by in vivo corticosterone treatment produces a 50% decrease of binding site number. Increases and decreases of binding site number were not associated with a significant modification of the affinity for corticosterone. The present data show that in vivo corticosterone modulates its own number of binding sites demonstrated by in vitro binding on brain sections, in a manner which is reminiscent of changes in cytosol receptors demonstrated by conventional biochemical methods. Thus, this in vitro method provides an alternative way to study the plasticity of hippocampal glucocorticoid receptors.

Pharmacology of an antiandrogen, anandron, used as an adjuvant therapy in the treatment of prostate cancer.

To improve the inhibition of prostate cancer growth obtained by surgical or chemical castration (estrogens or LHRH analogs), blockade of the action of residual androgens of adrenal origin has been proposed. Among antiandrogens acting through the androgen receptor (AR), the nonsteroid anandron (RU 23908) has several advantages over available compounds: megestrol acetate and cyproterone acetate, both steroids, bind substantially to other hormone receptors (progestin, gluco- and mineralocorticoid); and anandron binds only to AR. The nonsteroid flutamide is a prodrug converted to the active metabolite, hydroxyflutamide; anandron is well absorbed on oral administration of an active dose and intact compound disappears slowly from plasma. This may explain why, although in vitro anandron interacts very transiently with AR, in vivo a high level of untransformed anandron is present at the receptor site to induce its antiandrogenic activity. Animal experiments confirm that anandron can counteract the effect of adrenal androgens and inhibit the LHRH analog-induced initial increase in androgen ("flare-up"). Thus, in rats castrated either surgically or by buserelin or DES and supplemented with adrenal androgens (since endogenous adrenal secretion is very low in this species compared to man), anandron decreased prostate weight to control levels. The administration of buserelin to intact rats over 15 days resulted in a significant increase in prostate weight between Days 1 and 5. The addition of anandron to the buserelin inhibited this increase and, furthermore, led to a far greater decrease in prostate weight than that due to buserelin alone at 15 days, indicating a synergy of action.

[Cancer of the prostate: biologic bases for the use of an antiandrogen in its treatment]. / Cancer de la prostate: bases biologiques pour l'emploi d'un antiandrogène dans le traitement.

Although orchiectomy, estrogens and LHRH agonists suppress testicular androgens, they are without effect on adrenal androgens which are converted into dihydrotestosterone in the prostate. It is therefore necessary to develop substances able to block the action of all androgens, whatever their source, on target organs. The non-steroid, Anandron (RU 23908), when administered orally, gives rise to a high and sustained plasma level of intact compound that inhibits testosterone binding to its receptor. This inhibition, however, occurs not only in the prostate but also in the pituitary. The negative feedback action of androgens is thus inhibited by Anandron resulting in an increased secretion of testosterone and explaining the necessity of combining Anandron with castration (whether surgical or medical by an LHRH agonist). Anandron opposes, on the one hand, the action of adrenal androgens and, on the one other, of the testosterone surge that occurs during the early days of treatment with the LHRH analog. The efficacy of the combined treatment has been demonstrated experimentally. Clinical trials are presently underway.

Transient transcriptional inhibition of the transferrin gene by cyclic AMP.

Transferrin mRNA content and gene transcription rate were measured in the liver of rats submitted to iron overload or depletion, castration, treatment with sexual steroid hormones, glucagon and cyclic AMP. The influence of puberty in males and females and of pregnancy was also analysed. Glucagon and cyclic AMP reduced mRNA level by about 50% at the 12th hour of treatment and transferrin gene transcription by as much as 95% at the 30th minute of drug infusion, with a secondary increase of the transcription rate for a protracted treatment. None of the other hormones tested had any detectable effect on transferrin gene expression, the same being true for iron overload or depletion.

In vitro binding of tritiated glucocorticoids directly on unfixed rat brain sections.

We describe a new technique for measuring specific in vitro binding of tritiated adrenal steroids on unfixed cryostat brain sections. The specific binding of [3H]corticosterone represents about 70% of the initial binding. Kinetic studies show that specific binding for [3H]corticosterone reaches equilibrium after 15 min incubation at room temperature. Scatchard analysis of [3H]corticosterone in vitro binding gives a linear plot with an apparent dissociation constant (Kd) and a number of binding sites (Bmax) in the range of 10(-8) M and 100 fmol/mg protein, respectively. [3H]Dexamethasone binding under the same conditions gives a similar Kd and a Bmax of 55 fmol/mg protein. The order of potency for the relative binding affinity for [3H]corticosterone labeled sites is as follows: corticosterone greater than progesterone, dexamethasone, RU 38486 (a "pure" antiglucocorticoid), RU 26988 (a "pure" glucocorticoid), aldosterone greater than estradiol, testosterone. Anatomical studies reveal that sections at the level of the hippocampus bind more [3H]corticosterone and [3H]dexamethasone in vitro than more rostral sections taken at the level of the septum. Adrenalectomy increases the capacity of [3H]corticosterone to bind to these sites and perfusion of the brain to remove transcortin and other blood proteins does not modify [3H]corticosterone binding. We conclude that it is possible to measure in unfixed frozen brain sections glucocorticoid binding sites.

RU 38486: potent antiglucocorticoid activity correlated with strong binding to the cytosolic glucocorticoid receptor followed by an impaired activation.

In order to explain the potent antiglucocorticoid activity of RU 38486 and the absence of agonist effect in spite of its very strong interaction with the cytoplasmic glucocorticoid receptor (GR), we investigated the compound's ability to promote GR "activation" and nuclear translocation. We have compared the dissociation-rates of the "non-activated" (molybdate stabilized) and of the "activated" (25 degrees C pre-heated) GR complexes formed either with [3H]RU 38486 or with different tritiated glucocorticoid agonists. While agonists dissociated more slowly from the "activated" than from the "non-activated" complex, RU 38486 dissociated much faster from the "activated" than from the "native" receptor. This difference of activation was confirmed in a DNA-cellulose binding assay. The affinity of the "activated" RU 38486-GR complex for DNA was much lower than that of the dexamethasone-GR complex. Finally, the in vitro nuclear uptake of [3H]RU 38486 was compared with that of [3H]dexamethasone after incubation with thymus minces at 25 or 37 degrees C. A very weak or nearly undetectable level of specific uptake of [3H]RU 38486 was observed in purified nuclei, whatever the concentration or the time of incubation used. These observations suggest that while glucocorticoid agonists form with the non-activated receptor a complex able to be activated into a more stable form (lower k-1), RU 38486 interacts strongly with the non-activated receptor (impeding the binding of DM) but the complex is "transformed" by heat to a less stable form (higher k-1), unable to translocate properly into the nucleus in order to trigger a glucocorticoid response.