Aromatase is the enzyme that catalyzes the conversion of androgens to estrogens. Initial studies of its enzymatic activity and function took place in an environment focused on estrogen as a component of the birth control pill. At an early stage, investigators recognized that inhibition of this enzyme could have major practical applications for treatment of hormone-dependent breast cancer, alterations of ovarian and endometrial function, and treatment of benign disorders such as gynecomastia. Two general approaches ultimately led to the development of potent and selective aromatase inhibitors. One targeted the enzyme using analogs of natural steroidal substrates to work out the relationships between structure and function. The other approach initially sought to block adrenal function as a treatment for breast cancer but led to the serendipitous finding that a nonsteroidal P450 steroidogenesis inhibitor, aminoglutethimide, served as a potent but nonselective aromatase inhibitor. Proof of the therapeutic concept of aromatase inhibition involved a variety of studies with aminoglutethimide and the selective steroidal inhibitor, formestane. The requirement for even more potent and selective inhibitors led to intensive molecular studies to identify the structure of aromatase, to development of high-sensitivity estrogen assays, and to "mega" clinical trials of the third-generation aromatase inhibitors, letrozole, anastrozole, and exemestane, which are now in clinical use in breast cancer. During these studies, unexpected findings led investigators to appreciate the important role of estrogens in males as well as in females and in multiple organs, particularly the bone and brain. These studies identified the important regulatory properties of aromatase acting in an autocrine, paracrine, intracrine, neurocrine, and juxtacrine fashion and the organ-specific enhancers and promoters controlling its transcription. The saga of these studies of aromatase and the ultimate utilization of inhibitors as highly effective treatments of breast cancer and for use in reproductive disorders serves as the basis for this first Endocrine Reviews history manuscript.

4-Hydroxy-f-androstene-3,17-dione (4-OH-A) when tested at various concentrations was found to inhibit markedly the conversion of 4-andorstene-3,17-dione to estrogens inhuman placental and rat ovarian microsomes. To obtain evidence that estrogen biosynthesis could also be reduced in vivo with 4-OH-A, rats were treated sc at a dose level of 50 mg/kg body weight. After 3 h the ovarian veins were cannulated and blood collected. Estradiol concentrations in the plasma were reduced by 80% compared to control values during the proestrous surge and on Day 4 of pregnancy. 4-OH-A was also found to be effective in controlling estrogen-dependent reproductive and neoplastic processes. In rats treated from Day 2-7 of pregnancy, implantation of fertilized ova was completely prevented in some rats, while in others either implantation was delayed or the development of implants was retarded. 4-OH-A treatment of rats having estrogen-dependent breast tumors induced by 7,12-dimethylbenz(a)anthracene caused 80% of the tumors to regress significantly in 4 weeks of treatment; 42% of these regressed completely.

No abstract

The synthesis and biological evaluation of androstenedione derivatives as inhibitors of estrogen biosynthesis are described. The results show that 4-hydroxy analogues are among the most potent in vitro inhibitors of the series. Esterification of the 4-hydroxy steroids generally reduced activity. Further conjugation of the 3-keto 4-ene system to give 4-hydroxy-4,6-androstadiene-3,17-dione caused more rapid inactivation of aromatase in rat ovarian microsomes than 4-hydroxyandrostenedione. Some compounds exhibited differences in activity when tested for inhibition of human placental microsomes vs. rat ovarian microsomes. The 4-hydroxyandrostenedione derivatives and their nonbulky esters were generally more potent in vitro and in vivo inhibitors than other substituted steroids in the series. Several of the synthesized compounds markedly reduce (50-81%) estrogen levels in rats on proestrus and/or had antifertility action. To date, none of the compounds surpassed the in vivo inhibitory action of 4-hydroxy-4-androstene-3,17-dione or its 4-acetate derivative.

To elucidate the mechanism for oxidative attack in ring A resulting in the production of estrogens, the stereochemistry and cofactor requirements for aromatization of estr-4-ene-3,l7-dione (I) were determined and compared with the aromatization characteristics of androst-4-ene-3,17-dione (II). Incubations of [4-I4C]II stereoselectively tritiated at C-l/3 or C-la with the lO.OOOg supernatant of human term placenta and a reduced nicotinamide-adenine dinucleotide phosphate generating system showed that the 13hydrogen was lost in the conversion to estrone. Incubation of [4-14C-l-3H (83% 3)]I under similar conditions yielded estrone in which 82% of the tritium T A he conversion of androstenedione1 or testosterone to estrogens by human placental preparations requires two general steps: (1) hydroxylation and eventual elimination of the C-10 methyl group, and (2) oxidative attack resulting in double-bond introduction in ring A (Talalay, 1965). The available evidence suggests step 2 occurs after C-19 hydroxylation and before or during cleavage of the C-10,19 bond. No potential intermediate beyond 19-oxoandrostenedione has been isolated, and it is not known how desaturation occurs in ring A to produce estrogen.' In a study to elucidate the events occurring in ring A, leading to aromatization, we found evidence, presented in a preliminary communication, that the 13-equatorial hydrogen is removed. This suggested enzymatic participation or steric interaction of the C-19 group at C-13 (Morato et al., 1962). 19-Nor steroids also serve as estrogen precursors in placental and ovarian preparations