Biotransformation of 17 beta-hydroxy-11 beta-(4-dimethylaminophenyl)17 alpha-1-propynyl-estra-4,9-dien-3-one (RU486) in rat hepatoma variants

In vitro antiprogestational/antiglucocorticoid activity and progestin and glucocorticoid receptor binding of the putative metabolites and synthetic derivatives of CDB-2914, CDB-4124, and mifepristone

In determining the biological profiles of various antiprogestins, it is important to assess the hormonal and antihormonal activity, selectivity, and potency of their proximal metabolites. The early metabolism of mifepristone is characterized by rapid demethylation and hydroxylation. Similar initial metabolic pathways have been proposed for CDB-2914 (CDB: Contraceptive Development Branch of NICHD) and CDB-4124, and their putative metabolites have been synthesized. We have examined the functional activities and potencies, in various cell-based assays, and relative binding affinities (RBAs) for progesterone receptors (PR) and glucocorticoid receptors (GR) of the putative mono- and didemethylated metabolites of CDB-2914, CDB-4124, and mifepristone and of the 17alpha-hydroxy and aromatic A-ring derivatives of CDB-2914 and CDB-4124. The binding affinities of the monodemethylated metabolites for rabbit uterine PR and human PR-A and PR-B were similar to those of the parent compounds. Monodemethylated mifepristone bound to rabbit thymic GR with higher affinity than monodemethylated CDB-2914 or CDB-4124. T47D-CO cells were used to assess inhibition of R5020-stimulated endogenous alkaline phosphatase activity and transactivation of the PRE(2)-thymidine kinase (tk)-luciferase (LUC) reporter plasmid in transient transfections. The antiprogestational potency was as follows: mifepristone/CDB-2914/CDB-4124/monodemethylated metabolites (IC(50)'s approximately 10(-9)M) > aromatic A-ring derivatives (IC(50)'s approximately 10(-8)M) > didemethylated/17alpha-hydroxy derivatives (IC(50)'s approximately 10(-7)M). Antiglucocorticoid activity was determined by inhibition of dexamethasone-stimulated transcriptional activity in HepG2 cells. The mono- and didemethylated metabolites of CDB-2914 and CDB-4124 had less antiglucocorticoid activity (IC(50)'s approximately 10(-6)M) than monodemethylated mifepristone (IC(50) approximately 10(-8)M) or the other test compounds. At 10(-6)M in transcription assays, none of these compounds showed progestin agonist activity, whereas mifepristone and its monodemethylated metabolite manifested slight glucocorticoid agonist activity. The reduced antiglucocorticoid activity of monodemethylated CDB-2914 and CDB-4124 was confirmed in vivo by the thymus involution assay in adrenalectomized male rats. The aromatic A-ring derivatives-stimulated transcription of an estrogen-responsive reporter plasmid in MCF-7 and T47D-CO human breast cancer cells but were much less potent than estradiol. Taken together, these data suggest that the proximal metabolites of mifepristone, CDB-2914, and CDB-4124 contribute significantly to the antiprogestational activity of the parent compounds in vivo. Furthermore, the reduced antiglucocorticoid activity of CDB-2914 and CDB-4124 compared to mifepristone in vivo may be due in part to decreased activity of their putative proximal metabolites.

In vitro metabolism of mifepristone (RU-486) in rat, monkey and human hepatic S9 fractions: identification of three new mifepristone metabolites

1. In vitro metabolism of the antiprogestin drug mifepristone (RU-486) was studied after incubation with rat, monkey and human hepatic S9 fractions in the presence of an NADPH-generating system. 2. Unchanged mifepristone (approximately 45% of the sample(s) in rat; approximately 70% in monkey; approximately 65% in human) plus six metabolites, three known and three new, were profiled, quantified and tentatively identified on the basis of MS and MS/MS data. 3. The proposed metabolic pathways for mifepristone are proposed, and the two metabolic steps are (A) N-demethylation and (B) methyl oxidation. 4. Step A formed N-desmethyl mifepristone (M1) in major amounts (approximately 35% s in rat, 16% in monkey and human) and N,N-didesmethyl mifepristone (M2) in minor amounts (< 5% s in all species). Step B, or in conjunction with step A, produced four minor/trace metabolites, namely hydroxymethyl mifepristone (M3), hydroxymethyl M1 (M4), hydroxymethyl M2 (M5) and formyl mifepristone (M6).

Mechanism-based inhibition of rat liver microsomal diazepam C3-hydroxylase by mifepristone associated with loss of spectrally detectable cytochrome P450

Since initial studies with the steroids norethindrone and ethynylestradiol, reported by White and Muller-Eberhard in 1977 (Biochem. J. 166, 57-64), there has been continuing interest in xenobiotics that bear terminal or sub-terminal acetylenic groups which can cause catalysis-dependent inhibition of CYP monooxygenases associated either with loss of prosthetic group heme or protein adduct formation. Mifepristone is a synthetic steroid bearing a propyne substitution on carbon 17 and this suggested to us that it may act as a mechanism-based inhibitor of the CYP isoforms responsible for its metabolism. In human and rat liver, CYP3A isoforms have been implicated in mifepristone clearance and mifepristone administration to rats has also been shown to induce CYP3A enzymes and the associated diazepam C3-hydroxylase activity (Cheesman, Mason and Reilly, J. Steroid Biochem. Mol. Biol., 58, 1996, 447-454). With microsomes prepared from the livers of untreated female rats and others in which diazepam C3-hydroxylase has been induced, we show here that mifepristone can cause catalysis-dependent inhibition of this monooxygenase. In addition, incubation of microsomes with mifepristone in the presence, but not in the absence, of NADPH caused loss of spectrally detectable cytochrome P450. These results suggest that heme adduct formation may result from mifepristone metabolism by CYP3A monooxygenases which undergo self-catalysed irreversible inactivation with this drug as substrate. Since mifepristone administration in vivo is able also to cause induction of the synthesis of hepatic CYP3A apoprotein, mifepristone may have the potential in human medicine for complex interactions with other co-administered drugs which are also substrates for CYP3A monooxygenases.

Identification of CYP3A4 as the principal enzyme catalyzing mifepristone (RU 486) oxidation in human liver microsomes

Various complementary approaches were used to elucidate the major cytochrome P450 (CYP) enzyme responsible for mifepristone (RU 486) demethylation and hydroxylation in human liver microsomes: chemical and immunoinhibition of specific CYPs; correlation analyses between initial rates of mifepristone metabolism and relative immunodetectable CYP levels and rates of CYP marker substrate metabolism; and evaluation of metabolism by cDNA-expressed CYP3A4. Human liver microsomes catalyzed the demethylation of mifepristone with mean (+/-SD) apparent K(m) and Vmax values of 10.6 +/- 3.8 microM and 4920 +/- 1340 pmol/min/mg protein, respectively; the corresponding values for hydroxylation of the compound were 9.9 +/- 3.5 microM and 610 +/- 260 pmol/min/mg protein. Progesterone and midazolam (CYP3A4 substrates) inhibited metabolite formation by up to 77%. The CYP3A inhibitors gestodene, triacetyloleandomycin, and 17 alpha-ethynylestradiol inhibited mifepristone demethylation and hydroxylation by 70-80%; antibodies to CYP3A4 inhibited these reactions by approximately 82 and 65%, respectively. In a bank of human liver microsomes from 14 donors, rates of mifepristone metabolism correlated significantly with relative immunodetectable CYP3A levels, rates of midazolam 1'-and 4-hydroxylation and rates of erythromycin N-demethylation, marker CYP3A catalytic activities (all r2 > or = 0.85 and P < 0.001). No significant correlations were observed for analyses with relative immunoreactive levels or marker catalytic activities of CYP1A2, CYP2C9, CYP2C19, CYP2D6, or CYP2E1. Recombinant CYP3A4 catalyzed mifepristone demethylation and hydroxylation with apparent K(m) values 7.4 and 4.1 microM, respectively. Collectively, these data clearly support CYP3A4 as the enzyme primarily responsible for mifepristone demethylation and hydroxylation in human liver microsomes.

Effects of food deprivation and adrenalectomy on CYP3A induction by RU486 in female rats

We have studied the effects of food deprivation and adrenalectomy on the induction by RU486 of female rat liver microsomal CYP3A apoprotein, erythromycin N-demethylase and diazepam C3-hydroxylase activities. RU486 was a potent inducer of CYP3A apoprotein in intact animals and food deprivation enhanced this response. Food deprivation alone caused only weak CYP3A apoprotein induction suggesting a synergistic interaction in the regulation of protein expression. These results were reflected in the measurements of diazepam C3-hydroxylase activity. This confirms diazepam C3-hydroxylase as a useful and easily measured index of CYP3A monooxygenase content in female rat liver microsomes. Erythromycin N-demethylase did not show concordance with this pattern; this monooxygenase was much more strongly induced by food deprivation alone than by RU486 administration and, in addition, adrenalectomy abolished the induction response to food deprivation. The lack of correspondence between the apoprotein and erythromycin N-demethylase results suggests that non-CYP3A or novel, hitherto uncharacterized CYP3A isoforms may contribute to erythromycin N-demethylation in female rats. The close agreement between the results for CYP3A apoprotein and diazepam C3-hydroxylase indicates that although RU486 possesses a terminal acetylenic moeity it does not, at the dosages used here, cause mechanism-based inactivation of the CYP3A monooxygenase protein it induces. Current studies are directed to characterizing the particular CYP3A isoform(s) whose production is stimulated by RU486.