Inactivation of human placental aromatase by 6 alpha- and 6 beta-hydroperoxyandrostenedione

Probing the binding pocket of the active site of aromatase with 6-ether or 6-ester substituted androst-4-ene-3,17-diones and their diene and triene analogs

A series of 6-ester- (3 and 4) and 6-ether- (7 and 8) substituted androst-4-ene-3,17-diones (androstenediones) and their 1,4-diene analogs (5 and 6, and 9 and 10) as well as C6-substituted 4,6-diene and 1,4,6-triene steroids 11 and 12 were synthesized as aromatase inhibitors to gain insight into the structure-activity relationship between various substituents and inhibitory activity. All of the inhibitors synthesized blocked aromatase in a competitive manner. The inhibitory activities of all of the steroids, except for the 6beta-benzoates 4g and 6h and the 6beta-acetate 6a, were fairly effective to very powerful (K(i): 7.0-320 nM). The 6alpha-n-hexanoyloxy- and 6alpha-benzyloxyandrostenediones (3e and 7e) were the most potent inhibitors (K(i): 7.0 nM each). In the series of 4-ene and 1,4-diene steroids, the 6alpha-substituted steroids had higher affinity for the enzyme than the corresponding 6beta-isomers. In the 1,4-diene steroid series, 6beta-substituted steroids 6a, e, g, and 10a, b, e caused a time-dependent inactivation of aromatase, whereas their 6alpha-isomers 5 and 9 essentially did not. The ether-substituted 1,4,6-trienes 12 inactivated the enzyme in a time-dependent manner; in contrast, their 4,6-diene analogs 11 did not. The substrate androstenedione blocked the inactivation, but no significant effect of L-cysteine was observed. Based on molecular modeling with the PM3 method, along with the present inhibition and inactivation results, it is thought that both the steric effects of the 6-substituents as well as the electronic effects of the C-6 oxygen functions play a critical role in the binding of inhibitors to the active site of aromatase.

Aromatase inhibitors--mechanisms of steroidal inhibitors

Inhibition of aromatase has been an attractive approach for examining the roles of estrogen biosynthesis in various physiological or pathological processes. Effective aromatase inhibitors can serve as potential therapeutic agents for controlling estrogen-dependent diseases such as hormone-dependent breast cancer. Investigations on the development of aromatase inhibitors have therefore expanded greatly in the past two decades. Numerous steroidal agents have been developed that have high affinities for the aromatase enzyme complex and exhibit either competitive inhibition, irreversible inhibition, or mechanism-based (enzyme-activated) inhibition. Mechanism-based inhibitors have distinct advantages in drug design, since these inhibitors are highly enzyme specific, produce prolonged inhibition, and exhibit minimal toxicities. Examination of the structure-activity relationships of the numerous steroidal aromatase inhibitors suggest that the spacial requirements for interaction of agents with the active site of aromatase are very restrictive, permitting only small structural changes to be made on the A-ring and at C-19. Incorporation of small polar substituents at the C-4 position, such as a hydroxyl group, or addition of aryl functionalities at the 7 alpha-position of the steroid, are the exceptions, and inhibitors with such modifications exhibit enhanced affinity for the enzyme. Future investigations of steroidal aromatase inhibitors as probes of the active site of purified aromatase will provide valuable information on enzyme structure at the molecular level, will permit a more detailed examination of the mechanisms of inhibition, and will enhance the development of more specific and effective inhibitors for the treatment of estrogen-dependent breast cancer.

Biological characterization of 10-(2-propynyl) estr-4-ene-3, 17-dione (MDL 18,962), an enzyme-activated inhibitor of aromatase

MDL 18,962 was shown to be a highly specific, potent (Ki = 3-4 nM), enzyme-activated inhibitor of aromatase with minimal intrinsic endocrine properties. The affinity of MDL 18,962 was higher for human and baboon placental aromatase than for rhesus placental or rodent ovarian aromatase. These species differences necessitated the development of a novel model of peripheral aromatase utilizing human enzyme. Human choriocarcinoma trophoblast xenografts in athymic nude mice were used for pharmacologic and pharmacokinetic evaluation of MDL 18,962. The ED50 for inhibition of aromatase activity in these trophoblast tumors at 6 h post-treatment was 1.4 mg/kg, s.c. and 3.0 mg/kg, oral. Preliminary results indicated that the ED50 for inhibition of peripheral aromatization of androgen by MDL 18,962 in female baboons was 0.01 mg/kg, i.v. and 4 mg/kg, oral.

Aromatase inhibitors and benign prostatic hyperplasia

A growing amount of evidence implies that estrogens may play a role together with androgens in the genesis of benign prostatic hyperplasia (BPH) in man. We review here some of this evidence together with advances made in characterizing inhibitors of estrogen biosynthesis (aromatase inhibitors). It is proposed that aromatase inhibitors may find application in non-surgical treatment of BPH.

Studies of the inactivation of human placental aromatase by 17 alpha-ethynyl-substituted 10 beta-hydroperoxy and related 19-nor steroids

The inactivation of human placental aromatase by 17 alpha-ethynyl-10 beta-hydroperoxy-17 beta-hydroxy-4-estren-3-one (SCH 10015) was investigated. In either the presence or absence of added NADPH, SCH 10015 (Ki = 41 microM) caused a time-dependent loss of aromatase activity (e.g. 50% loss after 20 min with 20 microM SCH 10015). Evidence for the oxidation of an active site sulfhydryl group as the molecular basis for SCH 10015 inactivation is presented. The contraceptive 17 alpha-ethynyl-substituted 19-nor steroids, norethisterone (Ki = 48 microM) and norethynodrel (Ki = 38 microM), were evaluated and found not to inactivate aromatase, suggesting that the potential conversion of either compound to SCH 10015 did not occur to a significant extent in these microsomal incubations. It is speculated that the previously observed potent contraceptive effects of SCH 10015 may have been the result of irreversible inhibition of estrogen biosynthesis.

Purification and reconstitution properties of human placental aromatase. A cytochrome P-450-type monooxygenase

The hemoprotein component of human placental aromatase (estrogen synthetase) has been purified to a high degree of homogeneity by a combination of affinity and adsorption chromatography on aminohexyl-Sepharose, concanavalin-A-Sepharose, and hydroxyapatite. The monomeric form of the enzyme has an Mr of 55000 +/- 1000 as estimated by sodium dodecyl sulfate gel electrophoresis. Its absolute spectrum shows a high-spin Soret band at 394 nm while its reduced, CO-difference spectrum has a maximum at 447 +/- 1 nm. Full reconstitution of aromatase activity was obtained when it was recombined with a homogeneous preparation of the higher-Mr form of either human placental, or bovine hepatic NADPH-cytochrome P-450 reductase. Critical factors for purification of the very unstable, membrane-bound hemoprotein with good retention of activity were, besides the chromatographic sequence, the use of the zwitterionic detergent 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (Chaps) during the solubilization, and the stabilizing effect of the aromatase substrate, 4-androstene-3,17-dione, throughout the procedure. In the presence of NADPH, the reconstituted enzyme system smoothly aromatizes 19-oxoandrostenedione, 19-hydroxyandrostenedione and androstenedione in this order of reactivity. The same reconstituted system also aromatized testosterone, but it was inactive towards 19-norandrostenedione. Known cytochrome P-450 inhibitors decreased its activity. We conclude: (a) the terminal oxidase of human placental aromatase is indeed a cytochrome P-450-type monooxygenase; (b) the multistep aromatization reaction of C19 androstenes is catalyzed by a single enzyme; (c) aromatization of 19-norsteroids reported by other authors must be due to a different aromatase. Experimental data obtained with the reconstituted enzyme are fully compatible with the concept of a reaction mechanism for the aromatization sequence involving an all-trans, antiparallel elimination of the 19-methyl group, the 2 beta proton and the 1 alpha proton, rather than the 1 beta proton, as generally assumed.