- Email: [email protected]
Cytochrome P450 aromatase (CYP19) in the non-human primate brain: distribution, regulation, and functional significance
Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
Cytochrome P450 aromatase (CYP19) in the non-human primate brain: distribution, regulation, and functional significance夽 Charles E. Roselli∗ , John A. Resko Department of Physiology and Pharmacology, Oregon Health and Sciences University, 3181 SW Sam Jackson Park Road, Portland, OR 97201-3098, USA
Abstract In adult male primates, estrogens play a role in both gonadotropin feedback and sexual behavior. Inhibition of aromatization in intact male monkeys acutely elevates serum levels of luteinizing hormone, an effect mediated, at least partially, within the brain. High levels of aromatase (CYP19) are present in the monkey brain and regulated by androgens in regions thought to be involved in the central regulation of reproduction. Androgens regulate aromatase pretranslationally and androgen receptor activation is correlated with the induction of aromatase activity. Aromatase and androgen receptor mRNAs display both unique and overlapping distributions within the hypothalamus and limbic system suggesting that androgens and androgen-derived estrogens regulate complimentary and interacting genes within many neural networks. Long-term castrated monkeys, like men, exhibit an estrogen-dependent neural deficit that could be an underlying cause of the insensitivity to testosterone that develops in states of chronic androgen deficiency. Future studies of in situ estrogen formation in brain in the primate model are important for understanding the importance of aromatase not only for reproduction, but also for neural functions such as memory and cognition that appear to be modulated by estrogens. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Cytochrome P450 aromatase; CYP19; Monkey; Hypothalamus; Amygdala; LH; Testosterone; Estradiol; mRNA; In situ hybridization
1. Introduction Androgens are necessary for feedback regulation of gonadotropin secretion and for maintaining libido in males. For a long time, a connection between androgen and estrogen was not made because they produced their effects through two receptor systems and some of their actions were antagonistic. Later it was found that androgens also act as a prohormone for the local production of estrogens in target tissues. The aromatase enzyme complex (CYP19) has the unique ability to convert androgens to estrogen through a multistep enzymatic pathway [1]. Aromatase is present in a variety of tissues, including skin, adipose, bone, muscle, and brain, so that the relatively low serum levels of estrogens in males most likely do not accurately reflect their concentration in target cells and their importance for mediating local effects. Clinical experiments [2–5], as well as endocrine evaluations of men that have mutations in the genes for aromatase or estrogen receptor [6], demonstrate that estrogen deficiency, either induced or naturally occurring, results in chronic elevations of gonadotropins in serum. 夽 Proceedings of the Symposium: ‘Aromatase 2000 and the Third Generation’ (Port Douglas, Australia, 3–7 November 2000). ∗ Corresponding author. Tel.: +1-503-494-5837; fax: +1-503-494-4352. E-mail address: [email protected] (C.E. Roselli).
This observation, which is indicative of a disrupted negative feedback mechanism, leaves little doubt that estrogens, and by inference aromatase, are crucial for the feedback control of gonadotropin secretion in men. However, gonadotropins are secreted in an episodic manner that is determined by the integrated response of both the hypothalamus and anterior pituitary to circulating steroids as well as environmental cues. For ethical and experimental reasons, it is not possible to access the hypothalamus, pituitary, or portal circulation in humans by directly measuring GnRH secretion or discretely inhibiting aromatase activity. Although a recent carefully designed and executed clinical study [7] implicates both the hypothalamus and pituitary as sites of estrogen’s action in men, it is impossible to directly and unequivocally prove that central aromatization is part of this process in humans. For this reason, an experimental animal model whose reproductive system is closely related to Homo sapiens is needed to address questions regarding the local production of estrogens from androgen precursors in the feedback regulation of gonadotropins. Experiments on laboratory rats suggest that aromatization of androgens to estrogens is not required for the suppression of gonadotropin secretion in rodents; although in rats this pathway is required for the expression of reproductive behaviors [8,9]. In contrast, our research, and that of others, clearly establishes that negative feedback regulation of gonadotropins in non-human primates has
0960-0760/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 0 7 6 0 ( 0 1 ) 0 0 1 4 1 - 8
248
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
both an androgenic and an estrogenic component [10–13] and that the regulation of neural aromatase activity (AA) is of prime importance to this physiological process. The non-human primate has long been recognized as an important model for endocrinological investigations relevant to the human reproduction especially for studies of female reproduction, because of the similarities between the control of the ovarian cycle in women and female monkeys [14]. The present review summarizes studies on macaques performed in our laboratory that have helped clarify the role of aromatase and estrogens in the neuroendocrine regulation of male reproduction.
2. Evidence that estrogen and aromatase play important roles in negative feedback regulation of LH secretion in adult male rhesus monkeys In non-human primates, testosterone and LH are secreted on a diurnal schedule [15]. Greater quantities of these hormones are secreted in the evening than in the morning hours [16]. After castration, gonadotropin concentrations gradually increase in the systemic circulation of rhesus macaques, but can be suppressed if large amounts of exogenous testosterone are administered [10,17]. The earliest experimental indication that estrogens contribute to negative feedback regulation of gonadotropins in male monkeys came from steroid replacement studies following castration. Administration of physiological levels of testosterone and estradiol at the time of castration maintained serum LH levels at or below the preoperative control ranges [10,17]. Neither testosterone nor estradiol given alone was an effective mediator of negative feedback under these experimental conditions [10]. With the development of potent and effective aromatase inhibitors it became possible to address the question of whether, and to what extent, aromatization contributes to the LH regulation under physiological conditions. We found that treatment of adult male rhesus monkeys (Macaca mulatta) with the aromatase inhibitor 1,4,6-androstatriene-3,17-dione
(ATD) elevated serum LH and testosterone concentrations 3–5-fold [11] (Fig. 1). Concurrent treatment of ATD-treated monkeys with small quantities of estradiol abolished the stimulatory effect of ATD. During treatment with ATD alone, peripheral estradiol levels were reduced by 30% and hypothalamic aromatase activity was reduced 80–90%. ATD lacked significant anti-androgenic activity that could account for these results. We found subsequently that ATD or one of its metabolites exerts significant androgenic actions in castrated cynomolgus monkeys (M. fasicularis), however, this agonist activity cannot account for the stimulatory effect on LH and testosterone secretion observed in the gonad-intact animal [12]. This study did suggest, however, that feedback control of the rise in LH following castration is controlled differently than in the intact male, the latter having an estrogenic component whereas the former does not [12]. A new generation non-steroidal aromatase inhibitors have proven to be more potent and more selective than ATD [18]. Preliminary studies on monkeys [13] and men [19] indicate that Ciba–Geigy aromatase inhibitors CGS16949A (fadrozole) and CGS47645 increase testicular testosterone production 10-fold and that this effect is related to an increase in pituitary LH secretion. Thus, maintenance of normal LH and testosterone levels in adult male primates depends on estrogen action. Other evidence from our laboratory further supports the idea that feedback control of LH secretion in male rhesus monkeys is estrogen driven. In a recent unpublished study, we treated castrated monkeys (n = 4) and treated them immediately with dihydrotestosterone (DHT). This treatment resulted in serum levels of dihydrotestosterone of ∼7.0 ng/ml. Even in the presence of supra-physiological circulating levels of dihydrotestosterone, LH concentrations rose significantly after castration reaching castrate levels with 30 days (unpublished observations). Thus, dihydrotestosterone, a pure androgen not able to be aromatized, was ineffective in maintaining precastration levels of LH. These data imply that estrogen alone or estrogen plus androgen are a required component of the negative feedback loop for controlling gonadotropin secretion in male monkeys.
Fig. 1. Individual mean levels of estradiol, LH, and testosterone in six adult male rhesus macaques before and after 21 days of ATD-induced aromatase suppression. The group mean ± S.E.M. for each parameter is indicated by the filled boxes in each graph. ATD (−), baseline state, ATD (+), aromatase-inhibited state; asterisks, significant change from baseline (redrawn from [11] with permission).
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
249
3. Sites of steroid action on LH In the monkey, testosterone acts primarily, although probably not exclusively, on the brain to regulate LH secretion by suppressing the frequency of the GnRH pulse generator. This was first inferred from analysis of the frequency and amplitude of LH secretion in the peripheral circulation, but has not yet been confirmed by directly measuring GnRH secretion in males [20]. Thus, castration leads to an increased LH pulse frequency and amplitude, whereas testosterone replacement reduces LH pulse frequency [21]. In contrast to the castration response of LH in normal males, orchidectomy does not elicit increased LH secretion in monkeys made hypogonadotropic by lesioning the hypothalamus (i.e. the hypophysiotropic clamp model [22]) and given an invariant frequency of pulsatile GnRH replacement [20]. Neither testosterone nor dihydrotestosterone affected the response of primary pituitary cell cultures to hourly GnRH pulses, nor did they affect LHα and LHβ gene expression [23]. These result supports the conclusion that androgens act directly in the brain to modulate GnRH and consequently LH secretion in monkeys. In contrast to androgen, estrogen appears to regulate LH secretion by acting at both the hypothalamus and pituitary in the male monkey. Studies in the primate demonstrated that direct administration of estradiol into the hypothalamus or third ventricle suppresses LH secretion [24,25]. Both testosterone and estradiol regulate hypothalamic GnRH content. We found that GnRH concentrations in the medial basal hypothalamus of male monkeys were significantly reduced after castration coincident with the hypersecretion of LH [26]. On the other hand, treatment of adult castrated males with physiological doses of testosterone or estradiol sufficient to suppress the LH response to castration, maintained significantly greater concentrations of GnRH in the infundibulum-median eminence than were present in castrated control monkeys [27]. Estradiol also suppressed GnRH-induced LH secretion from male monkey pituitary cells in primary culture, indicating a pituitary site of action for estrogen negative feedback in the non-human primate [23]. Taken together these results suggest that in monkeys androgens act primarily to restrain GnRH pulsatile secretion, while estrogens acts both to inhibit the GnRH secretion and to suppress pituitary responsiveness to GnRH (Fig. 2). Future studies will need to determine the specific neural pathways and subcellular mechanism that underlie steroid negative feedback in male monkeys.
4. Sites of estrogen synthesis A substantial amount of estrogen is produced by both peripheral and central aromatization of androgens in vivo [28–30] making it difficult to determine the relative importance of these estrogen sources for the negative feedback regulation of LH secretion in male monkeys. However,
Fig. 2. Summary diagram of the negative (−) and positive (+) feedback actions of testicular steroids in male primates.
unequivocal evidence that central aromatization is involved in the control of pulsatile LH secretion has recently been obtained in male sheep [31]. Infusion of the aromatase inhibitor fadrozole into the third ventricle at doses that did not affect plasma estradiol were able to increase LH pulse frequency in rams, a result that is consistent with a feedback role for central aromatization. Similar in vivo studies have not been attempted in non-human primates. The ability of neural tissue from the non-human primate to aromatize androgens to estrogens was first demonstrated almost 30 years ago [32]. In order to identify the specific sites of aromatization in primate brain and understand the factors regulating its activity, we performed a series of experiments employing brain microdissection technology and a highly sensitive radiometric aromatase assay [26,33,34]. We found that aromatase activity is highest within specific regions of the hypothalamus and amygdala, i.e. the medial preoptic area-anterior hypothalamus (MPAH), ventromedial hypothalamus (VMH), bed nucleus of the stria terminalis (BNST), cortical amygdala (CA), and medial amygdala (MA). Levels of aromatase activity in the medial basal hypothalamus (MBH), the brain region thought to regulate tonic gonadotropin secretion [35], are significantly correlated with GnRH content [26]. Aromatase activity in MPAH declined significantly 6 weeks after castration [26], and increased after testosterone treatment in both MPAH and VMH [33]. We also found that androgen receptors are selectively activated by testosterone in the regions of the monkey brain where testosterone is effective at regulating aromatase [34]. The correlation between androgen receptor activation and aromatase induction suggests that testosterone acts through an androgen receptor-mediated mechanism to regulate aromatase activity, similar to what
250
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
we demonstrated for the rat [36]. In non-human primates, as well as several other vertebrate species, aromatase protein is located in both neuronal cell bodies and axon terminals where it may act through classical as well as non-classical steroid signaling pathways [37].
5. Distribution and regulation of aromatase mRNA in the monkey brain To further investigate neural aromatase, we examined the expression of aromatase mRNA within the rhesus monkey brain [38,39]. For our initial studies, we cloned a 455-nucleotide (nt) cDNA that spanned exons 3–5 of the coding region. Using this probe, we developed a ribonuclease protection assay and studied the distribution of aromatase mRNA in monkey brain. Two RNA fragments were protected by our assay, one was the expected full-length of 455-nt, and the other was truncated at 300-nt. Based on studies in the rat, we presume that the truncated transcript represents an alternate aromatase mRNA lacking the initial coding exons 2 and 3 [40–42], and presumably codes for a non-functional protein [43]. The quantitative distribution of 455-nt protected band was significantly correlated with the distribution of aromatase activity (correction coefficient = 0.895; P < 0.001). The highest levels of expression were observed in the BNST, amygdala, MPAH, and MBH, which includes VMH. In a second study, male rhesus monkeys were either castrated or castrated and treated with either testosterone or dihydrotestosterone in order to determine whether aromatase is regulated by androgens pretranslationally [39]. Aromatase mRNA levels in the MPAH and MBH of both androgen treatment groups were significantly higher than levels in the corresponding castrated controls. Other brain areas were not affected by castration and androgen treatment. Thus, androgens regulate aromatase pretranslationally within the brain areas thought to be important for the regulation of neuroendocrine functions (Table 1). Table 1 A comparison of the effect of castration and androgen treatment on brain AA and mRNA levels in the monkey braina Treatment
MPAH VMH BNST MA CA
Castrationb
Testosterone/DHTc
AA
AA
mRNA
↓ – – ↓ –
–/↑ ↑/↑ –/– –/– –/–
↑ ↑ – – –
a Androgens regulate aromatase pre-translationally in the MPAH and VMH. b Effect of castration for 6 weeks compared in intact controls [26]. c Effect of treatment for 3 weeks with physiological doses of testosterone and DHT compared to castrate controls [33,39].
However, we do not currently understood what factors regulate aromatase expression in other areas of the brain. This is an important question for future investigation because estrogens have been shown to have widespread beneficial effects on other neural functions besides the regulation of gonadotropin secretion and reproductive behavior, such as cognition and neuronal survival [44,45]. In a recent study, we employed in situ hybridization histochemistry to map the regional locations of cells expressing aromatase and androgen receptor mRNA in the adult male macaque hypothalamus and amygdala [46]. We utilized reverse transcriptase/polymerase chain reaction to synthesize a monkey-specific aromatase cDNA fragment that spans exons 2 and 3. This 254-nt cDNA (Aro254) was designed to recognize the initial coding region encoding for full-length aromatase mRNA and will not recognize alternate aromatase mRNA transcripts that lacks the first two coding exons [38,40–42]. The monkey-specific AR cDNA employed was cloned using a similar approach and spanned exons 6 through 8 in the steroid-binding domain [47]. Intense labeling of aromatase and androgen receptor mRNA-containing cells was observed in discrete hypothalamic areas including the medial preoptic nucleus and ventromedial hypothalamic nucleus. Strongly labeled aromatase and androgen receptor mRNA-containing cells were identified in the central component of the medial preoptic nucleus, a region that appears analogous to the sexually dimorphic preoptic nucleus in other species. Aromatase mRNA-containing neurons were most abundant in the rostral hypothalamus and preoptic area whereas androgen receptor mRNA-containing neurons were most numerous in the medial basal and posterior hypothalamus. Moderate to heavily labeled cells were present in telencephalic regions that have strong reciprocal inputs with the hypothalamus, i.e. the bed nucleus of the stria terminalis, lateral ventral septum, medial amygdala, cortical amygdala and basal medial amygdala. All areas that contained aromatase mRNA-containing cells also contained androgen receptor mRNA-expressing cells, but there were areas in which androgen receptor mRNA was expressed but not aromatase mRNA, i.e. median eminence, posterior hypothalamus, premammillary nucleus, and tuberomammillary nucleus. Our data demonstrate that aromatase and androgen receptor mRNAs display both unique and overlapping distributions within many regions of the hypothalamus and limbic system that serve essential roles in the central regulation of reproductive function. This distribution appears to be highly conserved in all mammalian species studied to date, for examples see [41,48]. Because these neural pathways have extensive interconnections, it is likely that androgens and, androgen-derived estrogens, regulate complimentary and interacting sets of genes within this distributed network of neurons. However, the apparent spatial differences in the expression of aromatase and androgen receptor mRNA-containing neurons suggest that androgenic steroids may act autonomously within specific neurons.
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
251
Fig. 3. Concentrations of aromatase activity, aromatase mRNA, and androgen receptor mRNA in MPAH and MBH of two groups of male rhesus macaques: one group was castrated and immediately treated with testosterone for 30 days (Cx + T 30 day); the other group was castrated and treated with testosterone for 9 days beginning 21 days after castration (Cx + T 9 day). Tissues were obtained on day 30 after Cx. Data are presented as mean ± S.E.M. (vertical lines). Bars marked with different letters differ significantly (P < 0.01) by Mann–Whitney U-tests (reprinted from [52] with permission).
6. Testosterone insensitivity and aromatase activity Male rhesus monkeys that have been castrated for more than 3 weeks are unresponsive to the negative feedback actions of testosterone [10,17,49]. Instituting a replacement regimen that produces circulating levels of testosterone that are within the physiologic range fails to suppress the hypersecretion of LH and FSH, unless estradiol is given concomitantly [10,17]. This apparent resistance to the inhibitory action of androgen on gonadotropin secretion in the chronic agonadal state has also been reported in castrated men and in men with primary gonadal failure [50,51]. Because central aromatization is an important source of bioactive estrogen in intact males, we reasoned that castration-induced androgen insensitivity in monkeys could result from impaired capacity for aromatization. To study this possibility, we compared aromatase activity, aromatase mRNA, and androgen receptor mRNA in two groups of male rhesus macaques: one group was castrated and immediately treated with testosterone for 30 days; the other group was castrated and treated with testosterone for 9 days beginning 21 days after castration [52]. Serum LH was suppressed at or below intact levels when testosterone replacement was initiated at the time of orchidectomy (Group 1), whereas when the same regimen of testosterone replacement was delayed for 3 weeks it was ineffective in suppressing LH (Group 2). These data suggest that Group 2 animals were resistant to the suppressive effect of testosterone on LH secretion. This state of androgen resistance was correlated with a decreased capacity for aromatization in specific brain regions, i.e. the medial preoptic-anterior hypothalamic region and medial basal hypothalamus (Fig. 3). Despite reduced levels of brain aromatase activity, the levels of aromatase mRNA and androgen receptor mRNA were
not altered in the Group 2 animals. Thus, we hypothesize that an estrogen-dependent neural deficit, not involving regulation of aromatase mRNA, occurs in long-term castrated monkeys.
7. Conclusions The central nervous system of the male monkey contains a heterogeneous distribution of neurons that can biosynthesize estrogen from androgen precursors supplied by the arterial system of the brain. Not only is androgen used as raw material for estrogen synthesis by these neurons but, it also exerts stimulatory effects on the aromatase mRNA, which serves as a template for production of the aromatase protein. Estrogens produced locally within the brain appear to be necessary components of feedback regulation of gonadotropin secretion in this species. In some parts of the brain, aromatase activity and its messenger RNA are not androgen-dependent but may have other unknown regulators. We speculate that a greater understanding of the mechanisms of in situ estrogen formation in the brain will lead to important new breakthroughs for understanding not only the reproductive system, but other neural functions such as cognition and memory as well. The use of non-human primates as models for these studies in lieu of using humans will be important.
Acknowledgements This work was supported, in part, by NIH grants HD18196 (JAR) and D43 TW HD00669 and NSF grant IBN 9817037 (CER).
252
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
References [1] E.R. Simpson, M.S. Mahendroo, G.D. Means, M.W. Kilgore, M.M. Hinshelwood, S.G. Lorence, B. Amarneh, Y. Ito, C.R. Fisher, M.D. Michael, C.R. Mendelson, S.E. Bulun, Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis, Endocrine Rev. 15 (1994) 342–355. [2] J.S. Finkelstein, L.S. O’Dea, R.W. Whitcomb, W.F. Crowley Jr., Sex steroid control of gonadotropin secretion in the human male. Part II. Effects of estradiol administration in normal and gonadotropin-releasing hormone-deficient men, J. Clin. Endocrinol. Metab. 73 (1991) 621–628. [3] J.S. Finkelstein, R.W. Whitcomb, L.S. O’Dea, C. Longcope, D.A. Schoenfeld, W.F. Crowley Jr., Sex steroid control of gonadotropin secretion in the human male. Part I. Effects of testosterone administration in normal and gonadotropin-releasing hormonedeficient men, J. Clin. Endocrinol. Metab. 73 (1991) 609–620. [4] S.J. Winters, J.J. Janick, D.L. Loriaux, R.J. Sherins, Studies on the role of sex steroids in the feedback control of gonadropin concentrations in men. Part II. Use of estrogen antagonist, clomiphene citrate, J. Clin. Endocrinol. Metab. 48 (1979) 222–227. [5] R.J. Santen, J.M. Leonard, R.J. Sherins, H.M. Gandy, C.A. Paulsen, Short- and long-term effects of clomiphene citrate on the pituitary–testicular axis, J. Clin. Endocrinol. 33 (1971) 970–979. [6] M. Faustini-Fustini, V. Rochira, C. Carani, Oestrogen deficiency in men: where are we today? Eur. J. Endocrinol. 140 (1999) 111–129. [7] F.J. Hayes, S.B. Seminara, S. Decruz, P.A. Boepple, W.F. Crowley Jr., Aromatase inhibition in the human male reveals a hypothalamic site of estrogen feedback, J. Clin. Endocrinol. Metab. 85 (2000) 3027–3035. [8] A.S. Bhatnagar, P. Müller, L. Schenkel, P.F. Trunet, I. Beh, K. Schieweck, Inhibition of estrogen biosynthesis and its consequences on gonadotrophin secretion in the male, J. Steroid Biochem. Mol. Biol. 41 (1992) 437–443. [9] A.N. Clancy, D. Zumpe, R.P. Michael, Intracerebral infusion of an aromatase inhibitor, sexual behavior and brain estrogen receptor-like immunoreactivity in intact male rats, Neuroendocrinology 61 (1995) 98–111. [10] J.A. Resko, S.K. Quadri, H.G. Spies, Negative feedback control of gonadotropins in male rhesus monkeys: effects of time after castration and interactions of testosterone and estradiol-17, Endocrinology 101 (1977) 215–224. [11] W.E. Ellinwood, D.L. Hess, C.E. Roselli, H.G. Spies, J.A. Resko, Inhibition of aromatization stimulates luteinizing hormone and testosterone secretion in adult male rhesus monkeys, J. Clin. Endocrinol. Metab. 59 (1984) 1088–1096. [12] J.A. Resko, P.B. Connolly, C.E. Roselli, S.E. Abdelgadir, J.V.A. Choate, Differential effects of aromatase inhibition on luteinizing hormone secretion in intact and castrated male cynomolgus macaques, J. Clin. Endocrinol. Metab. 77 (1993) 1529–1534. [13] G. Shetty, H. Krishnamurthy, H.N. Krishnamurthy, S. Bhatnagar, R.N. Moudgal, Effect of estrogen deprivation on the reproductive physiology of male and female primates, J. Steriod Biochem. Mol. Biol. 61 (1997) 166. [14] J. Hotchkiss, E. Knobil, The menstrual cycle and its neuroendocrine control, in: E. Knobil, J.D. Neill (Eds.), The Physiology of Reproduction, 2nd Edition, Raven, New York, 1994, pp. 711–749. [15] T.M. Plant, Time courses of concentration of circulating gonadotropin, prolactin, testosterone, and cortisol in adult male rhesus monkeys (Macaca mulatta) throughout the 24 h light–dark cycle, Biol. Reprod. 25 (1981) 244–252. [16] R.L. Goodman, J. Hotchkiss, F.J. Karsch, Diurnal variations in serum testosterone concentrations in the adult male rhesus monkey, Biol. Reprod. 11 (1974) 624–630. [17] T.M. Plant, D.L. Hess, J. Hotchkiss, E. Knobil, Testosterone and the control of gonadotropin secretion in the male rhesus monkey (Macaca mulatta), Endocrinology 103 (1978) 535–541.
[18] R.J. Santen, H.A. Harvey, Use of aromatase inhibitors in breast carcinoma, Endocr. Relat Cancer 6 (1999) 75–92. [19] P.F. Trunet, P. Mueller, A.S. Bhatnagar, I. Dickes, G. Monnet, G. White, Open dose-finding study of a new potent and selective nonsteroidal aromatase inhibitor, CGS20267, in healthy male subjects, J. Clin. Endocrinol. Metab. 77 (1993) 319–323. [20] T.M. Plant, Gonadal regulation of hypothalamic gonadotropinreleasing hormone release in primates, Endocrine Res. 7 (1986) 75– 88. [21] T.M. Plant, Effects of orchidectomy and testosterone replacement treatment on pulsatile luteinizing hormone secretion in the adult rhesus monkey (Macaca mulatta), Endocrinology 110 (1982) 1905– 1913. [22] E. Knobil, T.M. Plant, L. Wildt, P.E. Belchetz, G. Marshall, Control of the rhesus monkey menstrual cycle: permissive role of hypothalamic gonadotropin-releasing hormone, Science 207 (1980) 1371–1373. [23] S. Kawakami, S.J. Winters, Regulation of luteinizing hormone secretion and subunit messenger ribonucleic acid expression by gonadal steroids in perfused pituitary cells from male monkeys and rats, Endocrinology 140 (1999) 3587–3593. [24] S.C. Chappel, J.A. Resko, R.L. Norman, H.G. Spies, Studies in rhesus monkeys on the site where estrogen inhibits gonadotropins: delivery of 17 beta-estradiol to the hypothalamus and pituitary gland, J. Clin. Endocrinol. Metab. 52 (1981) 1–8. [25] M. Ferin, P.W. Carmel, E.A. Zimmerman, M. Warren, R. Perez, R.L. Vande Wiele, Location of intrahypothalamic estrogen-responsive sites influencing LH secretion in the female rhesus monkey, Endocrinology 95 (1974) 1059–1068. [26] C.E. Roselli, H. Stadelman, L.E. Horton, J.A. Resko, Regulation of androgen metabolism and luteinizing hormone-releasing hormone content in discrete hypothalamic and limbic areas of male rhesus macaques, Endocrinology 120 (1987) 97–106. [27] C.E. Roselli, J.A. Resko, Regulation of hypothalamic luteinizing hormone-releasing hormone levels by testosterone and estradiol in male rhesus monkeys, Brain Res. 509 (1990) 343–346. [28] C. Franz, C. Longcope, Androgen and estrogen metabolism in male rhesus monkeys, Endocrinology 105 (1979) 869–874. [29] C. Longcope, R.B. Billiar, Y. Takaoka, P.S. Reddy, D. Richardson, B. Little, Tissue sites of aromatization in the female rhesus monkey, Endocrinology 113 (1983) 1679–1682. [30] C. Martel, M.H. Melner, D. Gagne, J. Simard, F. Labrie, Widespread tissue distribution of steroid sulfatase, 3 beta-hydroxysteroid dehydrogenase/delta 5-delta 4 isomerase (3 beta-HSD), 17 beta-HSD 5 alpha-reductase and aromatase activities in the rhesus monkey, Mol. Cell Endocrinol. 104 (1994) 103–111. [31] T.P. Sharma, D. Blache, M.A. Blackberry, G.B. Martin, Role of peripheral and central aromatization in the control of gonadotrophin secretion in the male sheep, Reprod. Fertil. Dev. 11 (1999) 293–302. [32] F. Flores, F. Naftolin, K.J. Ryan, Aromatization of androstenedione and testosterone by rhesus monkey hypothalamus and limbic system, Neuroendocrinology 11 (1973) 117–182. [33] C.E. Roselli, J.A. Resko, Testosterone regulates aromatase activity in discrete brain areas of male rhesus macaques, Biol. Reprod. 40 (1989) 1–6. [34] J.A. Resko, P.B. Connolly, C.E. Roselli, S.E. Abdelgadir, J.V.A. Choate, Selective activation of androgen receptors in the subcortical brain of male cynomolgus macaques by physiological hormone levels and its relationship to androgen-dependent aromatase activity, J. Clin. Endocrinol. Metab. 76 (1993) 1588–1593. [35] R.L. Goodman, E. Knobil, The sites of action of ovarian steroids in the regulation of LH secretion, Neuroendocrinology 32 (1981) 57–63. [36] C.E. Roselli, J.A. Resko, Sex differences in androgen-regulated expression of cytochrome P450 aromatase in the rat brain, J. Steroid Biochem. Mol. Biol. 61 (1997) 365–374.
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253 [37] F. Naftolin, T.L. Horvath, R.L. Jakab, C. Leranth, N. Harada, J. Balthazart, Aromatase immunoreactivity in axon terminals of the vertebrate brain—an immunocytochemical study on quail, monkey and human tissues, Neuroendocrinology 63 (1996) 149–155. [38] S.E. Abdelgadir, C.E. Roselli, J.V.A. Choate, J.A. Resko, Distribution of aromatase cytochrome P450 messenger ribonucleic acid in adult rhesus monkey brains, Biol. Reprod. 57 (1997) 772–777. [39] J.A. Resko, A.C. Pereyra-Martinez, H.L. Stadelman, C.E. Roselli, Region-specific regulation of cytochrome P450 aromatase messenger ribonucleic acid by androgen in brains of male rhesus monkeys, Biol. Reprod. 62 (2000) 1818–1822. [40] S.E. Abdelgadir, J.A. Resko, S.R. Ojeda, E.D. Lephart, M.J. McPhaul, C.E. Roselli, Androgens regulate aromatase cytochrome P450 messenger ribonucleic acid in rat brain, Endocrinology 135 (1994) 395–401. [41] C.E. Roselli, S.E. Abdelgadir, O.K. Ronnekleiv, S.A. Klosterman, Anatomic distribution and regulation of aromatase gene expression in the rat brain, Biol. Reprod. 58 (1998) 79–87. [42] N. Yamada-Mouri, S. Hirata, J. Kato, The expression and distribution of the aromatase mRNA variant in the adult rat brain, Program of the 10th International Congress of Endocrinology, 874, 1996. [43] Y.-C. Kao, T. Higashiyama, C. Yarborough, Y. Osawa, S. Chen, Functional characterization of 102-amino acid-deleted form of human aromatase (D102-aromatase), Steroids 64 (1999) 422–429. [44] B. Sherwin, T. Tulandi, Add-back estrogen reverses cognitive deficits induced by a gonadotropin-releasing hormone agonist in women with leiomyomata uteri, J. Clin. Endocrinol. Metab. 81 (1996) 2545–2549. [45] J.A. Chowen, I. Torres-Alemán, L.M. Garcia-Segura, Trophic effects of estradiol on fetal rat hypothalamic neurons, Neuroendocrinology 56 (1992) 895–901.
253
[46] C.E. Roselli, S.A. Klosterman, J.A. Resko, Anatomic relationships between aromatase and androgen receptor mRNA expression in the hypothalamus and amygdala of adult male cynomolgus monkeys, J. Comp. Neurol. 439 (2001) 208–223. [47] S.E. Abdelgadir, C.E. Roselli, J.V.A. Choate, J.A. Resko, Androgen receptor messenger ribonucleic acid in brains and pituitaries of male rhesus monkeys: studies on distribution, hormonal control, and relationship to luteinizing hormone secretion, Biol. Reprod. 60 (1999) 1251–1256. [48] C.E. Roselli, F. Stormshak, J.A. Resko, Distribution of aromatase mRNA in the ram hypothalamus: an in situ hybridization study, J. Neuroendocrinol. 12 (2000) 656–664. [49] S.J. Winters, P. Troen, T.M. Plant, Relationship between testosterone binding globulin and the failure of androgens to suppress serum gonadotropin concentrations in long-term castrated adult male rhesus monkeys (Macaca mulatta), J. Steroid Biochem. 14 (1981) 1223– 1227. [50] S.J. Winters, R.J. Sherins, D.L. Loriaux, Studies on the role of sex steroids in the feedback control of gonadotropin concentrations in men. Part iii. Androgen resistance in primary gonadal failure, J. Clin. Endocrinol. Metab. 48 (1979) 553–558. [51] P.T. Capell, C.A. Paulsen, D. Derleth, R. Skoglund, S. Plymate, The effect of short-term testosterone administration on serum FSH, LH and testosterone levels: evidence for selective abnormality in LH control in patients with Klinefelter’s syndrome, J. Clin. Endocrinol. Metab. 37 (1973) 752–759. [52] J.A. Resko, A.C. Pereyra-Martinez, H.L. Stadelman, C.E. Roselli, Cellular observations and hormonal correlates of feedback control of luteinizing hormone secretion by testosterone in long-term castrated male rhesus monkeys, Biol. Reprod. 63 (2000) 872–878.
Cytochrome P450 aromatase (CYP19) in the non-human primate brain: distribution, regulation, and functional significance夽 Charles E. Roselli∗ , John A. Resko Department of Physiology and Pharmacology, Oregon Health and Sciences University, 3181 SW Sam Jackson Park Road, Portland, OR 97201-3098, USA
Abstract In adult male primates, estrogens play a role in both gonadotropin feedback and sexual behavior. Inhibition of aromatization in intact male monkeys acutely elevates serum levels of luteinizing hormone, an effect mediated, at least partially, within the brain. High levels of aromatase (CYP19) are present in the monkey brain and regulated by androgens in regions thought to be involved in the central regulation of reproduction. Androgens regulate aromatase pretranslationally and androgen receptor activation is correlated with the induction of aromatase activity. Aromatase and androgen receptor mRNAs display both unique and overlapping distributions within the hypothalamus and limbic system suggesting that androgens and androgen-derived estrogens regulate complimentary and interacting genes within many neural networks. Long-term castrated monkeys, like men, exhibit an estrogen-dependent neural deficit that could be an underlying cause of the insensitivity to testosterone that develops in states of chronic androgen deficiency. Future studies of in situ estrogen formation in brain in the primate model are important for understanding the importance of aromatase not only for reproduction, but also for neural functions such as memory and cognition that appear to be modulated by estrogens. © 2002 Elsevier Science Ltd. All rights reserved. Keywords: Cytochrome P450 aromatase; CYP19; Monkey; Hypothalamus; Amygdala; LH; Testosterone; Estradiol; mRNA; In situ hybridization
1. Introduction Androgens are necessary for feedback regulation of gonadotropin secretion and for maintaining libido in males. For a long time, a connection between androgen and estrogen was not made because they produced their effects through two receptor systems and some of their actions were antagonistic. Later it was found that androgens also act as a prohormone for the local production of estrogens in target tissues. The aromatase enzyme complex (CYP19) has the unique ability to convert androgens to estrogen through a multistep enzymatic pathway [1]. Aromatase is present in a variety of tissues, including skin, adipose, bone, muscle, and brain, so that the relatively low serum levels of estrogens in males most likely do not accurately reflect their concentration in target cells and their importance for mediating local effects. Clinical experiments [2–5], as well as endocrine evaluations of men that have mutations in the genes for aromatase or estrogen receptor [6], demonstrate that estrogen deficiency, either induced or naturally occurring, results in chronic elevations of gonadotropins in serum. 夽 Proceedings of the Symposium: ‘Aromatase 2000 and the Third Generation’ (Port Douglas, Australia, 3–7 November 2000). ∗ Corresponding author. Tel.: +1-503-494-5837; fax: +1-503-494-4352. E-mail address: [email protected] (C.E. Roselli).
This observation, which is indicative of a disrupted negative feedback mechanism, leaves little doubt that estrogens, and by inference aromatase, are crucial for the feedback control of gonadotropin secretion in men. However, gonadotropins are secreted in an episodic manner that is determined by the integrated response of both the hypothalamus and anterior pituitary to circulating steroids as well as environmental cues. For ethical and experimental reasons, it is not possible to access the hypothalamus, pituitary, or portal circulation in humans by directly measuring GnRH secretion or discretely inhibiting aromatase activity. Although a recent carefully designed and executed clinical study [7] implicates both the hypothalamus and pituitary as sites of estrogen’s action in men, it is impossible to directly and unequivocally prove that central aromatization is part of this process in humans. For this reason, an experimental animal model whose reproductive system is closely related to Homo sapiens is needed to address questions regarding the local production of estrogens from androgen precursors in the feedback regulation of gonadotropins. Experiments on laboratory rats suggest that aromatization of androgens to estrogens is not required for the suppression of gonadotropin secretion in rodents; although in rats this pathway is required for the expression of reproductive behaviors [8,9]. In contrast, our research, and that of others, clearly establishes that negative feedback regulation of gonadotropins in non-human primates has
0960-0760/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 0 7 6 0 ( 0 1 ) 0 0 1 4 1 - 8
248
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
both an androgenic and an estrogenic component [10–13] and that the regulation of neural aromatase activity (AA) is of prime importance to this physiological process. The non-human primate has long been recognized as an important model for endocrinological investigations relevant to the human reproduction especially for studies of female reproduction, because of the similarities between the control of the ovarian cycle in women and female monkeys [14]. The present review summarizes studies on macaques performed in our laboratory that have helped clarify the role of aromatase and estrogens in the neuroendocrine regulation of male reproduction.
2. Evidence that estrogen and aromatase play important roles in negative feedback regulation of LH secretion in adult male rhesus monkeys In non-human primates, testosterone and LH are secreted on a diurnal schedule [15]. Greater quantities of these hormones are secreted in the evening than in the morning hours [16]. After castration, gonadotropin concentrations gradually increase in the systemic circulation of rhesus macaques, but can be suppressed if large amounts of exogenous testosterone are administered [10,17]. The earliest experimental indication that estrogens contribute to negative feedback regulation of gonadotropins in male monkeys came from steroid replacement studies following castration. Administration of physiological levels of testosterone and estradiol at the time of castration maintained serum LH levels at or below the preoperative control ranges [10,17]. Neither testosterone nor estradiol given alone was an effective mediator of negative feedback under these experimental conditions [10]. With the development of potent and effective aromatase inhibitors it became possible to address the question of whether, and to what extent, aromatization contributes to the LH regulation under physiological conditions. We found that treatment of adult male rhesus monkeys (Macaca mulatta) with the aromatase inhibitor 1,4,6-androstatriene-3,17-dione
(ATD) elevated serum LH and testosterone concentrations 3–5-fold [11] (Fig. 1). Concurrent treatment of ATD-treated monkeys with small quantities of estradiol abolished the stimulatory effect of ATD. During treatment with ATD alone, peripheral estradiol levels were reduced by 30% and hypothalamic aromatase activity was reduced 80–90%. ATD lacked significant anti-androgenic activity that could account for these results. We found subsequently that ATD or one of its metabolites exerts significant androgenic actions in castrated cynomolgus monkeys (M. fasicularis), however, this agonist activity cannot account for the stimulatory effect on LH and testosterone secretion observed in the gonad-intact animal [12]. This study did suggest, however, that feedback control of the rise in LH following castration is controlled differently than in the intact male, the latter having an estrogenic component whereas the former does not [12]. A new generation non-steroidal aromatase inhibitors have proven to be more potent and more selective than ATD [18]. Preliminary studies on monkeys [13] and men [19] indicate that Ciba–Geigy aromatase inhibitors CGS16949A (fadrozole) and CGS47645 increase testicular testosterone production 10-fold and that this effect is related to an increase in pituitary LH secretion. Thus, maintenance of normal LH and testosterone levels in adult male primates depends on estrogen action. Other evidence from our laboratory further supports the idea that feedback control of LH secretion in male rhesus monkeys is estrogen driven. In a recent unpublished study, we treated castrated monkeys (n = 4) and treated them immediately with dihydrotestosterone (DHT). This treatment resulted in serum levels of dihydrotestosterone of ∼7.0 ng/ml. Even in the presence of supra-physiological circulating levels of dihydrotestosterone, LH concentrations rose significantly after castration reaching castrate levels with 30 days (unpublished observations). Thus, dihydrotestosterone, a pure androgen not able to be aromatized, was ineffective in maintaining precastration levels of LH. These data imply that estrogen alone or estrogen plus androgen are a required component of the negative feedback loop for controlling gonadotropin secretion in male monkeys.
Fig. 1. Individual mean levels of estradiol, LH, and testosterone in six adult male rhesus macaques before and after 21 days of ATD-induced aromatase suppression. The group mean ± S.E.M. for each parameter is indicated by the filled boxes in each graph. ATD (−), baseline state, ATD (+), aromatase-inhibited state; asterisks, significant change from baseline (redrawn from [11] with permission).
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
249
3. Sites of steroid action on LH In the monkey, testosterone acts primarily, although probably not exclusively, on the brain to regulate LH secretion by suppressing the frequency of the GnRH pulse generator. This was first inferred from analysis of the frequency and amplitude of LH secretion in the peripheral circulation, but has not yet been confirmed by directly measuring GnRH secretion in males [20]. Thus, castration leads to an increased LH pulse frequency and amplitude, whereas testosterone replacement reduces LH pulse frequency [21]. In contrast to the castration response of LH in normal males, orchidectomy does not elicit increased LH secretion in monkeys made hypogonadotropic by lesioning the hypothalamus (i.e. the hypophysiotropic clamp model [22]) and given an invariant frequency of pulsatile GnRH replacement [20]. Neither testosterone nor dihydrotestosterone affected the response of primary pituitary cell cultures to hourly GnRH pulses, nor did they affect LHα and LHβ gene expression [23]. These result supports the conclusion that androgens act directly in the brain to modulate GnRH and consequently LH secretion in monkeys. In contrast to androgen, estrogen appears to regulate LH secretion by acting at both the hypothalamus and pituitary in the male monkey. Studies in the primate demonstrated that direct administration of estradiol into the hypothalamus or third ventricle suppresses LH secretion [24,25]. Both testosterone and estradiol regulate hypothalamic GnRH content. We found that GnRH concentrations in the medial basal hypothalamus of male monkeys were significantly reduced after castration coincident with the hypersecretion of LH [26]. On the other hand, treatment of adult castrated males with physiological doses of testosterone or estradiol sufficient to suppress the LH response to castration, maintained significantly greater concentrations of GnRH in the infundibulum-median eminence than were present in castrated control monkeys [27]. Estradiol also suppressed GnRH-induced LH secretion from male monkey pituitary cells in primary culture, indicating a pituitary site of action for estrogen negative feedback in the non-human primate [23]. Taken together these results suggest that in monkeys androgens act primarily to restrain GnRH pulsatile secretion, while estrogens acts both to inhibit the GnRH secretion and to suppress pituitary responsiveness to GnRH (Fig. 2). Future studies will need to determine the specific neural pathways and subcellular mechanism that underlie steroid negative feedback in male monkeys.
4. Sites of estrogen synthesis A substantial amount of estrogen is produced by both peripheral and central aromatization of androgens in vivo [28–30] making it difficult to determine the relative importance of these estrogen sources for the negative feedback regulation of LH secretion in male monkeys. However,
Fig. 2. Summary diagram of the negative (−) and positive (+) feedback actions of testicular steroids in male primates.
unequivocal evidence that central aromatization is involved in the control of pulsatile LH secretion has recently been obtained in male sheep [31]. Infusion of the aromatase inhibitor fadrozole into the third ventricle at doses that did not affect plasma estradiol were able to increase LH pulse frequency in rams, a result that is consistent with a feedback role for central aromatization. Similar in vivo studies have not been attempted in non-human primates. The ability of neural tissue from the non-human primate to aromatize androgens to estrogens was first demonstrated almost 30 years ago [32]. In order to identify the specific sites of aromatization in primate brain and understand the factors regulating its activity, we performed a series of experiments employing brain microdissection technology and a highly sensitive radiometric aromatase assay [26,33,34]. We found that aromatase activity is highest within specific regions of the hypothalamus and amygdala, i.e. the medial preoptic area-anterior hypothalamus (MPAH), ventromedial hypothalamus (VMH), bed nucleus of the stria terminalis (BNST), cortical amygdala (CA), and medial amygdala (MA). Levels of aromatase activity in the medial basal hypothalamus (MBH), the brain region thought to regulate tonic gonadotropin secretion [35], are significantly correlated with GnRH content [26]. Aromatase activity in MPAH declined significantly 6 weeks after castration [26], and increased after testosterone treatment in both MPAH and VMH [33]. We also found that androgen receptors are selectively activated by testosterone in the regions of the monkey brain where testosterone is effective at regulating aromatase [34]. The correlation between androgen receptor activation and aromatase induction suggests that testosterone acts through an androgen receptor-mediated mechanism to regulate aromatase activity, similar to what
250
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
we demonstrated for the rat [36]. In non-human primates, as well as several other vertebrate species, aromatase protein is located in both neuronal cell bodies and axon terminals where it may act through classical as well as non-classical steroid signaling pathways [37].
5. Distribution and regulation of aromatase mRNA in the monkey brain To further investigate neural aromatase, we examined the expression of aromatase mRNA within the rhesus monkey brain [38,39]. For our initial studies, we cloned a 455-nucleotide (nt) cDNA that spanned exons 3–5 of the coding region. Using this probe, we developed a ribonuclease protection assay and studied the distribution of aromatase mRNA in monkey brain. Two RNA fragments were protected by our assay, one was the expected full-length of 455-nt, and the other was truncated at 300-nt. Based on studies in the rat, we presume that the truncated transcript represents an alternate aromatase mRNA lacking the initial coding exons 2 and 3 [40–42], and presumably codes for a non-functional protein [43]. The quantitative distribution of 455-nt protected band was significantly correlated with the distribution of aromatase activity (correction coefficient = 0.895; P < 0.001). The highest levels of expression were observed in the BNST, amygdala, MPAH, and MBH, which includes VMH. In a second study, male rhesus monkeys were either castrated or castrated and treated with either testosterone or dihydrotestosterone in order to determine whether aromatase is regulated by androgens pretranslationally [39]. Aromatase mRNA levels in the MPAH and MBH of both androgen treatment groups were significantly higher than levels in the corresponding castrated controls. Other brain areas were not affected by castration and androgen treatment. Thus, androgens regulate aromatase pretranslationally within the brain areas thought to be important for the regulation of neuroendocrine functions (Table 1). Table 1 A comparison of the effect of castration and androgen treatment on brain AA and mRNA levels in the monkey braina Treatment
MPAH VMH BNST MA CA
Castrationb
Testosterone/DHTc
AA
AA
mRNA
↓ – – ↓ –
–/↑ ↑/↑ –/– –/– –/–
↑ ↑ – – –
a Androgens regulate aromatase pre-translationally in the MPAH and VMH. b Effect of castration for 6 weeks compared in intact controls [26]. c Effect of treatment for 3 weeks with physiological doses of testosterone and DHT compared to castrate controls [33,39].
However, we do not currently understood what factors regulate aromatase expression in other areas of the brain. This is an important question for future investigation because estrogens have been shown to have widespread beneficial effects on other neural functions besides the regulation of gonadotropin secretion and reproductive behavior, such as cognition and neuronal survival [44,45]. In a recent study, we employed in situ hybridization histochemistry to map the regional locations of cells expressing aromatase and androgen receptor mRNA in the adult male macaque hypothalamus and amygdala [46]. We utilized reverse transcriptase/polymerase chain reaction to synthesize a monkey-specific aromatase cDNA fragment that spans exons 2 and 3. This 254-nt cDNA (Aro254) was designed to recognize the initial coding region encoding for full-length aromatase mRNA and will not recognize alternate aromatase mRNA transcripts that lacks the first two coding exons [38,40–42]. The monkey-specific AR cDNA employed was cloned using a similar approach and spanned exons 6 through 8 in the steroid-binding domain [47]. Intense labeling of aromatase and androgen receptor mRNA-containing cells was observed in discrete hypothalamic areas including the medial preoptic nucleus and ventromedial hypothalamic nucleus. Strongly labeled aromatase and androgen receptor mRNA-containing cells were identified in the central component of the medial preoptic nucleus, a region that appears analogous to the sexually dimorphic preoptic nucleus in other species. Aromatase mRNA-containing neurons were most abundant in the rostral hypothalamus and preoptic area whereas androgen receptor mRNA-containing neurons were most numerous in the medial basal and posterior hypothalamus. Moderate to heavily labeled cells were present in telencephalic regions that have strong reciprocal inputs with the hypothalamus, i.e. the bed nucleus of the stria terminalis, lateral ventral septum, medial amygdala, cortical amygdala and basal medial amygdala. All areas that contained aromatase mRNA-containing cells also contained androgen receptor mRNA-expressing cells, but there were areas in which androgen receptor mRNA was expressed but not aromatase mRNA, i.e. median eminence, posterior hypothalamus, premammillary nucleus, and tuberomammillary nucleus. Our data demonstrate that aromatase and androgen receptor mRNAs display both unique and overlapping distributions within many regions of the hypothalamus and limbic system that serve essential roles in the central regulation of reproductive function. This distribution appears to be highly conserved in all mammalian species studied to date, for examples see [41,48]. Because these neural pathways have extensive interconnections, it is likely that androgens and, androgen-derived estrogens, regulate complimentary and interacting sets of genes within this distributed network of neurons. However, the apparent spatial differences in the expression of aromatase and androgen receptor mRNA-containing neurons suggest that androgenic steroids may act autonomously within specific neurons.
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
251
Fig. 3. Concentrations of aromatase activity, aromatase mRNA, and androgen receptor mRNA in MPAH and MBH of two groups of male rhesus macaques: one group was castrated and immediately treated with testosterone for 30 days (Cx + T 30 day); the other group was castrated and treated with testosterone for 9 days beginning 21 days after castration (Cx + T 9 day). Tissues were obtained on day 30 after Cx. Data are presented as mean ± S.E.M. (vertical lines). Bars marked with different letters differ significantly (P < 0.01) by Mann–Whitney U-tests (reprinted from [52] with permission).
6. Testosterone insensitivity and aromatase activity Male rhesus monkeys that have been castrated for more than 3 weeks are unresponsive to the negative feedback actions of testosterone [10,17,49]. Instituting a replacement regimen that produces circulating levels of testosterone that are within the physiologic range fails to suppress the hypersecretion of LH and FSH, unless estradiol is given concomitantly [10,17]. This apparent resistance to the inhibitory action of androgen on gonadotropin secretion in the chronic agonadal state has also been reported in castrated men and in men with primary gonadal failure [50,51]. Because central aromatization is an important source of bioactive estrogen in intact males, we reasoned that castration-induced androgen insensitivity in monkeys could result from impaired capacity for aromatization. To study this possibility, we compared aromatase activity, aromatase mRNA, and androgen receptor mRNA in two groups of male rhesus macaques: one group was castrated and immediately treated with testosterone for 30 days; the other group was castrated and treated with testosterone for 9 days beginning 21 days after castration [52]. Serum LH was suppressed at or below intact levels when testosterone replacement was initiated at the time of orchidectomy (Group 1), whereas when the same regimen of testosterone replacement was delayed for 3 weeks it was ineffective in suppressing LH (Group 2). These data suggest that Group 2 animals were resistant to the suppressive effect of testosterone on LH secretion. This state of androgen resistance was correlated with a decreased capacity for aromatization in specific brain regions, i.e. the medial preoptic-anterior hypothalamic region and medial basal hypothalamus (Fig. 3). Despite reduced levels of brain aromatase activity, the levels of aromatase mRNA and androgen receptor mRNA were
not altered in the Group 2 animals. Thus, we hypothesize that an estrogen-dependent neural deficit, not involving regulation of aromatase mRNA, occurs in long-term castrated monkeys.
7. Conclusions The central nervous system of the male monkey contains a heterogeneous distribution of neurons that can biosynthesize estrogen from androgen precursors supplied by the arterial system of the brain. Not only is androgen used as raw material for estrogen synthesis by these neurons but, it also exerts stimulatory effects on the aromatase mRNA, which serves as a template for production of the aromatase protein. Estrogens produced locally within the brain appear to be necessary components of feedback regulation of gonadotropin secretion in this species. In some parts of the brain, aromatase activity and its messenger RNA are not androgen-dependent but may have other unknown regulators. We speculate that a greater understanding of the mechanisms of in situ estrogen formation in the brain will lead to important new breakthroughs for understanding not only the reproductive system, but other neural functions such as cognition and memory as well. The use of non-human primates as models for these studies in lieu of using humans will be important.
Acknowledgements This work was supported, in part, by NIH grants HD18196 (JAR) and D43 TW HD00669 and NSF grant IBN 9817037 (CER).
252
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253
References [1] E.R. Simpson, M.S. Mahendroo, G.D. Means, M.W. Kilgore, M.M. Hinshelwood, S.G. Lorence, B. Amarneh, Y. Ito, C.R. Fisher, M.D. Michael, C.R. Mendelson, S.E. Bulun, Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis, Endocrine Rev. 15 (1994) 342–355. [2] J.S. Finkelstein, L.S. O’Dea, R.W. Whitcomb, W.F. Crowley Jr., Sex steroid control of gonadotropin secretion in the human male. Part II. Effects of estradiol administration in normal and gonadotropin-releasing hormone-deficient men, J. Clin. Endocrinol. Metab. 73 (1991) 621–628. [3] J.S. Finkelstein, R.W. Whitcomb, L.S. O’Dea, C. Longcope, D.A. Schoenfeld, W.F. Crowley Jr., Sex steroid control of gonadotropin secretion in the human male. Part I. Effects of testosterone administration in normal and gonadotropin-releasing hormonedeficient men, J. Clin. Endocrinol. Metab. 73 (1991) 609–620. [4] S.J. Winters, J.J. Janick, D.L. Loriaux, R.J. Sherins, Studies on the role of sex steroids in the feedback control of gonadropin concentrations in men. Part II. Use of estrogen antagonist, clomiphene citrate, J. Clin. Endocrinol. Metab. 48 (1979) 222–227. [5] R.J. Santen, J.M. Leonard, R.J. Sherins, H.M. Gandy, C.A. Paulsen, Short- and long-term effects of clomiphene citrate on the pituitary–testicular axis, J. Clin. Endocrinol. 33 (1971) 970–979. [6] M. Faustini-Fustini, V. Rochira, C. Carani, Oestrogen deficiency in men: where are we today? Eur. J. Endocrinol. 140 (1999) 111–129. [7] F.J. Hayes, S.B. Seminara, S. Decruz, P.A. Boepple, W.F. Crowley Jr., Aromatase inhibition in the human male reveals a hypothalamic site of estrogen feedback, J. Clin. Endocrinol. Metab. 85 (2000) 3027–3035. [8] A.S. Bhatnagar, P. Müller, L. Schenkel, P.F. Trunet, I. Beh, K. Schieweck, Inhibition of estrogen biosynthesis and its consequences on gonadotrophin secretion in the male, J. Steroid Biochem. Mol. Biol. 41 (1992) 437–443. [9] A.N. Clancy, D. Zumpe, R.P. Michael, Intracerebral infusion of an aromatase inhibitor, sexual behavior and brain estrogen receptor-like immunoreactivity in intact male rats, Neuroendocrinology 61 (1995) 98–111. [10] J.A. Resko, S.K. Quadri, H.G. Spies, Negative feedback control of gonadotropins in male rhesus monkeys: effects of time after castration and interactions of testosterone and estradiol-17, Endocrinology 101 (1977) 215–224. [11] W.E. Ellinwood, D.L. Hess, C.E. Roselli, H.G. Spies, J.A. Resko, Inhibition of aromatization stimulates luteinizing hormone and testosterone secretion in adult male rhesus monkeys, J. Clin. Endocrinol. Metab. 59 (1984) 1088–1096. [12] J.A. Resko, P.B. Connolly, C.E. Roselli, S.E. Abdelgadir, J.V.A. Choate, Differential effects of aromatase inhibition on luteinizing hormone secretion in intact and castrated male cynomolgus macaques, J. Clin. Endocrinol. Metab. 77 (1993) 1529–1534. [13] G. Shetty, H. Krishnamurthy, H.N. Krishnamurthy, S. Bhatnagar, R.N. Moudgal, Effect of estrogen deprivation on the reproductive physiology of male and female primates, J. Steriod Biochem. Mol. Biol. 61 (1997) 166. [14] J. Hotchkiss, E. Knobil, The menstrual cycle and its neuroendocrine control, in: E. Knobil, J.D. Neill (Eds.), The Physiology of Reproduction, 2nd Edition, Raven, New York, 1994, pp. 711–749. [15] T.M. Plant, Time courses of concentration of circulating gonadotropin, prolactin, testosterone, and cortisol in adult male rhesus monkeys (Macaca mulatta) throughout the 24 h light–dark cycle, Biol. Reprod. 25 (1981) 244–252. [16] R.L. Goodman, J. Hotchkiss, F.J. Karsch, Diurnal variations in serum testosterone concentrations in the adult male rhesus monkey, Biol. Reprod. 11 (1974) 624–630. [17] T.M. Plant, D.L. Hess, J. Hotchkiss, E. Knobil, Testosterone and the control of gonadotropin secretion in the male rhesus monkey (Macaca mulatta), Endocrinology 103 (1978) 535–541.
[18] R.J. Santen, H.A. Harvey, Use of aromatase inhibitors in breast carcinoma, Endocr. Relat Cancer 6 (1999) 75–92. [19] P.F. Trunet, P. Mueller, A.S. Bhatnagar, I. Dickes, G. Monnet, G. White, Open dose-finding study of a new potent and selective nonsteroidal aromatase inhibitor, CGS20267, in healthy male subjects, J. Clin. Endocrinol. Metab. 77 (1993) 319–323. [20] T.M. Plant, Gonadal regulation of hypothalamic gonadotropinreleasing hormone release in primates, Endocrine Res. 7 (1986) 75– 88. [21] T.M. Plant, Effects of orchidectomy and testosterone replacement treatment on pulsatile luteinizing hormone secretion in the adult rhesus monkey (Macaca mulatta), Endocrinology 110 (1982) 1905– 1913. [22] E. Knobil, T.M. Plant, L. Wildt, P.E. Belchetz, G. Marshall, Control of the rhesus monkey menstrual cycle: permissive role of hypothalamic gonadotropin-releasing hormone, Science 207 (1980) 1371–1373. [23] S. Kawakami, S.J. Winters, Regulation of luteinizing hormone secretion and subunit messenger ribonucleic acid expression by gonadal steroids in perfused pituitary cells from male monkeys and rats, Endocrinology 140 (1999) 3587–3593. [24] S.C. Chappel, J.A. Resko, R.L. Norman, H.G. Spies, Studies in rhesus monkeys on the site where estrogen inhibits gonadotropins: delivery of 17 beta-estradiol to the hypothalamus and pituitary gland, J. Clin. Endocrinol. Metab. 52 (1981) 1–8. [25] M. Ferin, P.W. Carmel, E.A. Zimmerman, M. Warren, R. Perez, R.L. Vande Wiele, Location of intrahypothalamic estrogen-responsive sites influencing LH secretion in the female rhesus monkey, Endocrinology 95 (1974) 1059–1068. [26] C.E. Roselli, H. Stadelman, L.E. Horton, J.A. Resko, Regulation of androgen metabolism and luteinizing hormone-releasing hormone content in discrete hypothalamic and limbic areas of male rhesus macaques, Endocrinology 120 (1987) 97–106. [27] C.E. Roselli, J.A. Resko, Regulation of hypothalamic luteinizing hormone-releasing hormone levels by testosterone and estradiol in male rhesus monkeys, Brain Res. 509 (1990) 343–346. [28] C. Franz, C. Longcope, Androgen and estrogen metabolism in male rhesus monkeys, Endocrinology 105 (1979) 869–874. [29] C. Longcope, R.B. Billiar, Y. Takaoka, P.S. Reddy, D. Richardson, B. Little, Tissue sites of aromatization in the female rhesus monkey, Endocrinology 113 (1983) 1679–1682. [30] C. Martel, M.H. Melner, D. Gagne, J. Simard, F. Labrie, Widespread tissue distribution of steroid sulfatase, 3 beta-hydroxysteroid dehydrogenase/delta 5-delta 4 isomerase (3 beta-HSD), 17 beta-HSD 5 alpha-reductase and aromatase activities in the rhesus monkey, Mol. Cell Endocrinol. 104 (1994) 103–111. [31] T.P. Sharma, D. Blache, M.A. Blackberry, G.B. Martin, Role of peripheral and central aromatization in the control of gonadotrophin secretion in the male sheep, Reprod. Fertil. Dev. 11 (1999) 293–302. [32] F. Flores, F. Naftolin, K.J. Ryan, Aromatization of androstenedione and testosterone by rhesus monkey hypothalamus and limbic system, Neuroendocrinology 11 (1973) 117–182. [33] C.E. Roselli, J.A. Resko, Testosterone regulates aromatase activity in discrete brain areas of male rhesus macaques, Biol. Reprod. 40 (1989) 1–6. [34] J.A. Resko, P.B. Connolly, C.E. Roselli, S.E. Abdelgadir, J.V.A. Choate, Selective activation of androgen receptors in the subcortical brain of male cynomolgus macaques by physiological hormone levels and its relationship to androgen-dependent aromatase activity, J. Clin. Endocrinol. Metab. 76 (1993) 1588–1593. [35] R.L. Goodman, E. Knobil, The sites of action of ovarian steroids in the regulation of LH secretion, Neuroendocrinology 32 (1981) 57–63. [36] C.E. Roselli, J.A. Resko, Sex differences in androgen-regulated expression of cytochrome P450 aromatase in the rat brain, J. Steroid Biochem. Mol. Biol. 61 (1997) 365–374.
C.E. Roselli, J.A. Resko / Journal of Steroid Biochemistry & Molecular Biology 79 (2001) 247–253 [37] F. Naftolin, T.L. Horvath, R.L. Jakab, C. Leranth, N. Harada, J. Balthazart, Aromatase immunoreactivity in axon terminals of the vertebrate brain—an immunocytochemical study on quail, monkey and human tissues, Neuroendocrinology 63 (1996) 149–155. [38] S.E. Abdelgadir, C.E. Roselli, J.V.A. Choate, J.A. Resko, Distribution of aromatase cytochrome P450 messenger ribonucleic acid in adult rhesus monkey brains, Biol. Reprod. 57 (1997) 772–777. [39] J.A. Resko, A.C. Pereyra-Martinez, H.L. Stadelman, C.E. Roselli, Region-specific regulation of cytochrome P450 aromatase messenger ribonucleic acid by androgen in brains of male rhesus monkeys, Biol. Reprod. 62 (2000) 1818–1822. [40] S.E. Abdelgadir, J.A. Resko, S.R. Ojeda, E.D. Lephart, M.J. McPhaul, C.E. Roselli, Androgens regulate aromatase cytochrome P450 messenger ribonucleic acid in rat brain, Endocrinology 135 (1994) 395–401. [41] C.E. Roselli, S.E. Abdelgadir, O.K. Ronnekleiv, S.A. Klosterman, Anatomic distribution and regulation of aromatase gene expression in the rat brain, Biol. Reprod. 58 (1998) 79–87. [42] N. Yamada-Mouri, S. Hirata, J. Kato, The expression and distribution of the aromatase mRNA variant in the adult rat brain, Program of the 10th International Congress of Endocrinology, 874, 1996. [43] Y.-C. Kao, T. Higashiyama, C. Yarborough, Y. Osawa, S. Chen, Functional characterization of 102-amino acid-deleted form of human aromatase (D102-aromatase), Steroids 64 (1999) 422–429. [44] B. Sherwin, T. Tulandi, Add-back estrogen reverses cognitive deficits induced by a gonadotropin-releasing hormone agonist in women with leiomyomata uteri, J. Clin. Endocrinol. Metab. 81 (1996) 2545–2549. [45] J.A. Chowen, I. Torres-Alemán, L.M. Garcia-Segura, Trophic effects of estradiol on fetal rat hypothalamic neurons, Neuroendocrinology 56 (1992) 895–901.
253
[46] C.E. Roselli, S.A. Klosterman, J.A. Resko, Anatomic relationships between aromatase and androgen receptor mRNA expression in the hypothalamus and amygdala of adult male cynomolgus monkeys, J. Comp. Neurol. 439 (2001) 208–223. [47] S.E. Abdelgadir, C.E. Roselli, J.V.A. Choate, J.A. Resko, Androgen receptor messenger ribonucleic acid in brains and pituitaries of male rhesus monkeys: studies on distribution, hormonal control, and relationship to luteinizing hormone secretion, Biol. Reprod. 60 (1999) 1251–1256. [48] C.E. Roselli, F. Stormshak, J.A. Resko, Distribution of aromatase mRNA in the ram hypothalamus: an in situ hybridization study, J. Neuroendocrinol. 12 (2000) 656–664. [49] S.J. Winters, P. Troen, T.M. Plant, Relationship between testosterone binding globulin and the failure of androgens to suppress serum gonadotropin concentrations in long-term castrated adult male rhesus monkeys (Macaca mulatta), J. Steroid Biochem. 14 (1981) 1223– 1227. [50] S.J. Winters, R.J. Sherins, D.L. Loriaux, Studies on the role of sex steroids in the feedback control of gonadotropin concentrations in men. Part iii. Androgen resistance in primary gonadal failure, J. Clin. Endocrinol. Metab. 48 (1979) 553–558. [51] P.T. Capell, C.A. Paulsen, D. Derleth, R. Skoglund, S. Plymate, The effect of short-term testosterone administration on serum FSH, LH and testosterone levels: evidence for selective abnormality in LH control in patients with Klinefelter’s syndrome, J. Clin. Endocrinol. Metab. 37 (1973) 752–759. [52] J.A. Resko, A.C. Pereyra-Martinez, H.L. Stadelman, C.E. Roselli, Cellular observations and hormonal correlates of feedback control of luteinizing hormone secretion by testosterone in long-term castrated male rhesus monkeys, Biol. Reprod. 63 (2000) 872–878.