Aromatase in the Central Nervous System1

Central (central nervous system and pituitary) aromatization appears to be a fundamental process for endocrine control and development. Metabolism of androgens to estrogens and the subsequent metabolism of estrogens have been proven in many species, including humans, and linked to estrogen action. Thus, aromatization appears to initiate or to be involved in activities of importance to endocrine function at the central level and their effects peripherally. In the context of breast cancer, central aromatization relates to the control of gonadotrophins and other pituitary-brain hor mones which may effect metabolism at the level of the breast. For example, follicle-stimulating hormone can increase aro matization and may be a factor in the control of such metabo lism in breast tissue. Aromatization by Central Tissues Aromatization of androgens by CMS3 tissues (Chart 1) has been proven to exist in all species studied thus far (4, 13, 21 ). The levels of activity are relatively modest compared to tissues classically known as estrogen-producing sources, such as placenta and ovary. However, the evidence for in vitro and in vivo CNS aromatization is firm and includes localization to areas of neuroendocrine activity, particularly the hypothalamus and limbic system. While the anterior pituitary gland has oc casionally been shown to form measurable estrogen in vitro, this is less consistent. It is now established that CNS aromatase activity is found in the microsomal compartment, occurs in areas known to contain estrogen receptors, and responds to sex steroids and other endocrine changes. Coupled with evi dence of other CNS steroidal and nonsteroidal hormone pro duction, these findings contribute to our acceptance of the hypothalamus and its linked limbic structures as endocrine tissues. The proof of CNS aromatization rests upon 3 types of evidence: direct in vitro studies; direct in vivo studies; and indirect in vivo studies. These data have been reviewed previ ously, and we will only touch upon details used to tie the biochemical findings to some of the effects of central (CNS and pituitary) aromatization. Early animal studies led us to the conclusion that one wellknown "androgen action," sexual differentiation of the rodent brain (10, 17), required local estrogen formation (22). This "aromatization hypothesis" led to direct in vitro testing of human fetal CNS tissues for their ability to aromatize andro gens. Positive results were obtained using the method that Kenneth Ryan's laboratory had developed previously for stud' Presented at the Conference "Aromatase: New Perspectives for Breast Cancer." December 6 to 9, 1981, Key Biscayne, Fla. This work was supported by NIH Grant HD13587. National Institute of Child Health and Human Develop ment. 2 Fellow of the Alfred P. Sloan Foundation. 3 The abbreviations used are: CNS. central nervous system; ERN, cell nuclear estrogen receptor: PRC. cylosol progestin receptor; LH. luteinizing hormone. ies on placental aromatase. Evidence of further metabolism of the formed estrogen by the brain was also adduced in these early experiments, since the addition of large amounts of unlabeled estrogen to the incubation (estrogen trap) resulted in the retention of more of the newly formed estrogen from the androgen substrate (21). Subsequent studies have shown this metabolism of estrogens to include hydroxylation at positions 2, 4, or 16 and indicate possible sources of varying estrogen agonist-antagonist, neurotransmitter-effective, mutagenic com pounds (16). These in vitro methods have been applied by us and others to confirm and localize central estrogen formation. Among the animals studied, adult and perinatal rats were shown to have the ability to form estrogen, this evidence confirming the initial hypothesis regarding the basis of rodent sexual differentiation. Direct in vivo evidence of this activity was furnished in rhesus monkeys during collaborative experiments with Professor R. White (21). Perfusion of the isolated monkey brain-pituitary with radiolabeled androgens resulted in the formation and localized retention of estrogen. Studies showing the selective uptake of newly formed estrogen following administration of radiolabeled androgen to whole animals are also consistent with central aromatization in vivo (14, 27). Control of Central Aromatization Experiments, largely in the adult rabbit, revealed that central aromatization is responsive to endocrine and, perhaps, to constitutional factors. Activity is increased by castration and exposure to androgens and estrogen. Progestins diminish aro matization by CNS tissues. Central tissues from males regularly have greater aromatization activity than do identical tissues from females (16). Callard ef al. (4) showed an increase in in vitro aromatization by turtle CNS with time of incubation. Canick et al. (5) and Vaccaro ef al. (26) showed diminution of aroma tase activity by cultured hypothalamic neurons under the influ ence of norepinephrine and isoproterenol. The progestational antiandrogen cyproterone acetate when given prior to dissec tion curtails in vitro CNS aromatization (14). No evidence of product inhibition of CNS aromatization by estrogens has been reported. Several groups have shown a change in aromatiza tion during the pre-postnatal period (8, 21 ); however, the cause of these age-related changes is unknown. The anatomic loca tion of CNS aromatase has been beautifully mapped by microdissection techniques (24). Implications of Central Aromatization Central aromatization forms a critical link between biochem ical and anatomic aspects of endocrine regulation. The effects of this in situ steroidogenesis are local, but they are amplified by subsequent changes in control of the target organs of the pituitary and of the brain itself. Estrogens (and androgens, probably acting via central aromatization) are important in the 3274s CANCER RESEARCH VOL. 42 on April 13, 2017. © 1982 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from