Testosterone Regulates the Activity and Expression of Aromatase in the Canary Neostriatum

Transcription

1 Testosterone Regulates the Activity and Expression of Aromatase in the Canary Neostriatum Leonida Fusani, 1 John B. Hutchison, 2 Manfred Gahr 3 1 Research Centre for Ornithology of the Max-Planck-Society, Andechs, Germany 2 The Babraham Institute, CB1 42T Cambridge, United Kingdom 3 Department of Developmental Neurobiology, Vrije Universiteit Amsterdam, 1081 HV Amsterdam, The Netherlands Received 2 March 2001; accepted 6 May 2001 ABSTRACT: The estrogen synthesizing enzyme, P450 aromatase, plays a critical role in the regulation of vertebrate sexual behavior. Songbirds differ from other avian species in the distribution and expression of aromatase in the telencephalon. The highest concentration of aromatase in the songbird brain is found in the caudomedial neostriatum (NCM). This area surrounds the only nucleus of the neural song system that contains estrogen receptors, the high vocal center (HVC). It has been suggested that estrogen produced in NCM via aromatization of circulating testosterone (T) is involved in song development and adult song plasticity. The modalities of regulation of aromatase in NCM are not well understood, and some studies suggest that in NCM, unlike in the preoptic-hypothalamic areas, aromatase is not regulated by androgen and/or estrogen. In this work, we studied whether the treatment of female canaries with T, which induces the development of malelike song and the masculinization of the song system, also induces an increase in the expression and activity of aromatase in NCM. Our results show that both the expression and activity of aromatase in NCM increase in female canaries following T treatment. This study provides the first direct evidence that T regulates telencephalic aromatase in songbirds, and suggests that an increase in estrogen production in NCM might be functional in neural and behavioral plasticity during phases of song organization John Wiley & Sons, Inc. J Neurobiol 49: 1 8, 2001 Keywords: testosterone; aromatase; NCM; canary; song system INTRODUCTION The enzyme P450 aromatase, which converts androgen into estrogen, regulates testosterone (T)-dependent sexual differentiation and control of sexual behavior in vertebrates (reviewed by Lephart, 1996; Balthazart and Ball, 1998). In songbirds, that is, species belonging to the subgroup Oscine of the order Passeriformes, the highest level of aromatase activity and expression in the brain are found in the caudomedial neostriatum (NCM) (Vockel et al., 1990a; Correspondence to: L. Fusani (fusani@erl.ornithol.mpg.de) John Wiley & Sons, Inc. Shen et al., 1995; Saldanha and Schlinger, 1997; Foidart et al., 1998; Metzdorf et al., 1999; Soma et al., 1999; Fusani et al., 2000; Silverin et al., 2000). This pattern is a peculiarity of songbirds (Metzdorf et al., 1999; Silverin et al., 2000) and contrasts with the typical pattern of nonsongbirds, in which high concentrations of aromatase are found in limbic areas such as the hypothalamic-preoptic areas (HPOA), but little or no aromatase is present in the neostriatum (Hutchison and Steimer, 1984; Schumacher and Balthazart, 1987; Schlinger and Callard, 1989). In the only sub-oscine Passerine investigated so far, the golden-collared manakin (Manacus vitellinus), there is a substantial amount of aromatase activity in the 1

2 2 Fusani et al. NCM, although the highest activity is found in the HPOA (Saldanha et al., 2000a). The expression of aromatase in the NCM of songbirds appears to be related to the control and/or perception of song. NCM receives auditory projections and is involved in song discrimination (reviewed in Margoliash, 1997), and estrogen formed in this region could be functional to memorization of species-specific song patterns (Schlinger, 1997). In addition, the region of NCM with the highest concentration of aromatase literally surrounds the neostriatal vocal control nucleus high vocal center (HVC) (Metzdorf et al., 1999; Fusani et al., 2000). This nucleus is the only region of the neural song system that contains estrogen receptors (Gahr et al., 1993; Bernard et al., 1999). Therefore, estrogen formed in NCM could be transported to HVC and control estrogen-dependent vocal functions. In the HPOA of birds and mammals aromatase is regulated by T or by its metabolite estradiol (E2) (Steimer and Hutchison, 1981; Roselli and Resko, 1984; Balthazart et al., 1994). At present, it is unclear whether androgen and/or estrogen regulate aromatase in the songbird NCM. Aromatase activity in NCM does not differ between sexes in the zebra finch (Vockel et al., 1990a) or in the brown-headed cowbird (Molothrus ater) (Saldanha and Schlinger, 1997). However, the number of aromatase-immunoreactive fibers in NCM is sexually dimorphic in zebra finches (Saldanha et al., 2000b). In male zebra finches, castration does not reduce significantly aromatase activity in NCM (Vockel et al., 1990b). Other studies have examined the correlation between NCM aromatase activity and/or expression and circulating androgen levels in several species, providing contrasting results. In male domestic canaries (Serinus canaria), aromatase expression in NCM is positively correlated with seasonal changes in the circulating levels of androgen and estrogen (Fusani et al., 2000). Similarly, aromatase activity in NCM decreases across the breeding season in the Lapland longspur (Calcarius lapponicus) in correlation with a decrease in the plasma levels of T (Soma et al., 1999). In another species, the pied flycatcher (Ficedula hypoleuca), NCM aromatase activity does not change during the breeding season, whereas HPOA aromatase activity does change (Foidart et al., 1998). Finally, recent work on four songbird species [chaffinch (Fringilla coelebs), willow warbler (Phylloscopus trochilus), great tit (Parus major), and pied flycatcher] has shown that, contrary to HPOA, aromatase activity in NCM is not correlated with T within and across species (Silverin et al., 2000). Thus, the seasonal correlation between aromatase and T found in the Lapland longspur and the canary (Soma et al., 1999; Fusani et al., 2000) may reflect general physiological changes rather than indicating a causal relationship. The aim of this work was to study whether T regulates aromatase expression and activity in the canary NCM. In songbirds castration often results in high, and highly variable, levels of circulating androgen and estrogen (Heid et al., 1985; Marler et al., 1987; Adkins-Regan et al., 1990) because the complete removal of gonadal tissue is difficult (Marler et al., 1987) and androgen can be secreted by extragonadal sources. Therefore, the classical approach of castration and hormone replacement is of difficult application in songbirds. As an alternative approach we used intact female canaries, which have low circulating levels of androgen (Weichel et al., 1986). We studied whether T treatment induces an increase in aromatase activity and expression in NCM of female canaries. Our results show that both the expression of the aromatase mrna and the activity of the enzyme in NCM are increased after T treatment, providing the first direct demonstration of an androgen-dependent regulation of aromatase in the songbird NCM. METHODS Animals and Experimental Design Domestic canaries were bought from local breeders in the autumn preceding the experiments, and kept in heterosexual groups in large aviaries. During this period, the animals were exposed to natural light conditions (South Germany, 48 north). At the end of March, 16 females were moved to individual cages and kept under a light cycle simulating the local photoperiod (12 14 h light/day). Eight females were implanted subcutaneously with one time-release pellet containing 1.5 mg of T (60-day release, 25 g/day; Innovative Research of America, Sarasota, FL). The pellets were implanted subcutaneously on the back of the bird. The remaining eight females were used as controls. Four weeks after the implantation, between 12:00 and 16:00, the birds were bled from the jugular vein using a 1-mL syringe and a heparinized needle and then killed by decapitation. The brain was quickly removed, halved sagittally, frozen on liquid nitrogen, and stored at 80 C until dissection. The blood was centrifuged at 2000 rpm for 5 min and the plasma was collected and stored at 80 C until assayed. The weight of the birds and of the ovary were recorded. Implants were checked regularly, and they were all in place at the end of the experiments. We used 60-day release pellets to ensure that no implant was empty before the experiments had been concluded.

3 Aromatase Regulation in the Canary Neostriatum 3 Song Behavior In the last 2 weeks of the experiment, we monitored the vocal behavior of the birds for min every day. Vocalizations were recorded with a W6 Professional Walkman tape recorder (Sony Corporation, Tokyo, Japan) and an AT9450 directional microphone (Audio-Technica Ltd., Leeds, UK). The acquisition and the analysis of vocalizations were carried out with a digital sound analysis system (Signal 3.0; Engineering Design, Belmont, MA). Spectrograms of songs were examined for the song structure, that is, for the presence of long repetitions of syllables ( tours ) which are typical of male song (Nottebohm and Nottebohm, 1978; Güttinger, 1979) and of female T-induced song (Weichel et al., 1986). Radioimmunoassay of Androgen and Estrogen Androstenedione (AE), 5 -dihydrotestosterone (DHT), T, estrone (E1), and E2 were measured by radioimmunoassay as described in detail previously (Fusani et al., 2000). Antisera were obtained from Endocrine Sciences (Tarzana, CA) and labeled steroids from NEN Life Science Products (Boston, MA). The average recoveries were between 72 and 81%. The detection limits (pg/ml plasma) of the hormone radioimmunoassays were as follows: AE, 64; DHT, 68; T, 30; E2, 16; E1, 84. The intra-assay variation was 8% for all assays. Interassay variation is not applicable because all the samples were run in a single assay. When a hormone was nondetectable in a sample, in the statistical comparison, the lower detection limit of the respective assay (adjusted for extraction loss) was used. Aromatase Activity Assay The left half of the brain was cut in parasagittal sections approximately 0.5 mm thick on a freezing plate at a temperature between 18 and 13 C. Samples of the NCM and the cerebellum were dissected from the frozen slices using a microscalpel with the help of a dissection microscope, using a topographic procedure (Fig. 1) based on the distribution of aromatase in the canary brain (Metzdorf et al., 1999). The cerebellum was dissected as a control area with low levels of activity (Fusani et al., 2000). In order to minimize enzyme inactivation, the dissected microsamples were immediately refrozen on dry ice and stored at 80 C until assayed. The activity of aromatase in dissected samples was measured by means of an in vitro assay based on the stereospecific release of tritium from the 1 -position of the androgen substrate, validated for the avian brain (Steimer and Hutchison, 1989). The intra-assay variation, calculated from six replicate determinations of pooled ring dove hypothalamic samples, was 5%. The interassay variation of the assay, determined with the same pool, was 10%. Figure 1 The schematic drawing shows a parasagittal section ( mm from the midline) of the canary brain and the location of the areas dissected for the measurement of aromatase activity in vitro. cer, cerebellum; HV, hyperstriatum ventrale; NC, caudal neostriatum; NCM, caudomedial neostriatum; v, lateral ventricle. In Situ Hybridization The right half of the brain was cut into 20 m parasagittal slices in a cryostat at 18 C, mounted onto Superfrost Plus slides (Fisher, Fairlawn, NJ), and stored at 80 C until processing. Probes were prepared using sequences for canary aromatase cloned in our laboratory and validated for the canary brain (Metzdorf et al., 1999). The synthesis and labeling of the probes with 35 S-CTP (1250 Ci/mmol; NEN Life Science Products) was performed using the Riboprobe System (Promega, Madison, WI) according to the manufacturer s instructions. Our in situ hybridization procedure is a modification of that described by Whitfield et al. (1990), and has been described in detail previously (Gahr and Metzdorf, 1997; Metzdorf et al., 1999). Sections were counterstained with the Nissl-stain thionin and observed under light- and dark-field illumination with a Leica microscope. The brains had been number coded before sectioning. Therefore, the observer had no knowledge about the identity of the animals during measurement. The areas to be analyzed were digitized with an image analysis system (Imatec, Munich, Germany) via a video camera connected to the microscope. The level of expression of aromatase-mrna was measured in the region of NCM surrounding the medial part of HVC, mm lateral. This region contains the highest concentrations of aromatase-mrna in male canaries (Metzdorf et al., 1999; Fusani et al., 2000). At this level, the area of aromatase expression extends from the border of HVC to about 1 mm ventrorostral from the lateral ventricle (Metzdorf et al., 1999; Fusani et al., 2000) (Fig. 2). We sampled a squared area of m near to HVC, that is, 0.2 mm rostral to the lateral ventricle (caudal NCM) (Fig. 2). In addition,

4 4 Fusani et al. we sampled an area of the same size located 0.8 mm rostrally to the lateral ventricle (rostral NCM), that is, at the rostral end of the region of NCM that expresses aromatase (Metzdorf et al., 1999; Fusani et al., 2000). The level of Figure 3 Plasma levels (mean S.E.M.) of testosterone (T), 5 -dihydrotestosterone (DHT), and 17 -estradiol (E2) in untreated female canaries and in females implanted for 4 weeks with a T-releasing pellet (25 g/day; T-treated). The dotted lines represent the detection limit of the radioimmunoassay for each hormone. *p.002, Mann-Whitney U test. aromatase-mrna expression was calculated as the percent of the sampled area covered by silver grains, corrected for background labeling by subtracting the area covered by silver grains in a region of the same section lacking specific labeling. The method used for the measurements has been published in detail previously (Fusani et al., 2000). RESULTS Figure 2 (A) Schematic drawing of a parasagittal section of the canary brain. (B) Darkfield microphotograph of the area indicated by the square in (A) stained for aromatasemrna. Intense labeling is found in the caudomedial neostriatum (NCM) and hippocampus (HP). The area delimited by the dotted line is the medial portion of the high vocal center (HVC), which lacks aromatase-mrna staining. The expression level of aromatase-mrna was measured in NCM in the area indicated by the white square, 0.2 mm rostrally to the lateral ventricle (indicated by the arrow), and in another area of the same size 0.8 mm rostrally to the later ventricle (not shown). Top is dorsal, right is caudal. The bar in (B) is 0.2 mm. cer, cerebellum; LH, lamina hyperstriatica; LMD, lamina medullaris dorsalis; LPO, lobus paraolfactorius; v, lateral ventricle. The T treatment successfully increased the plasma levels of androgen and estrogen (Fig. 3). T, DHT, and E2 were detectable in all T-treated females and undetectable in all control females (Fig. 3). These differences were highly significant when compared with a Mann-Whitney test (Fig. 3). In T-treated females, the plasma levels of T, DHT, and E2 were in the range of those found in male canaries kept in seminatural conditions in the spring (Fusani et al., 2000). AE and E1 were undetectable in all samples (respectively: 64 and 84 pg/ml plasma), and were therefore excluded from further analyses. The ovary was regressed in both groups (T-treated: mg; control: mg) and did not differ between groups (t test, NS). All T-treated females developed malelike song within 2 4 weeks of implantation, that is, the songs were mainly composed of distinct phrases formed by continuous repetitions of syllables. No songlike vocalizations were recorded from untreated females, which produced only short calls. Data about song development in T-treated females will be published in detail elsewhere. The aromatase activity in NCM was three times

5 Aromatase Regulation in the Canary Neostriatum 5 Figure 4 Aromatase activity (means S.E.M.) in the cerebellum (cer) and caudomedial neostriatum (NCM) of untreated female canaries and of females implanted for 4 weeks with a T-releasing pellet (25 g/day; T-treated). *p.005, t test. higher in T-treated females than in controls, and the difference was highly significant (Fig. 4). In T-treated females, the levels of aromatase activity in NCM were in the range observed in male canaries in the spring (Fusani et al., 2000). In the cerebellum, the aromatase activity was low and was not affected by the T treatment (Fig. 4). The expression level of aromatase-mrna in NCM was measured in eight T-treated and six control females. The T treatment significantly increased the expression level of aromatase-mrna in the caudal portion of NCM (Fig. 5). On the contrary, there were no significant differences in aromatase-mrna expression in the rostral NCM (Fig. 5). In T-treated females the aromatase-mrna expression level in the caudal NCM was significantly correlated with the plasma levels of T (r S 0.738, n 8, p.05). T was not correlated with aromatase-mrna expression in the rostral NCM (r S 0.214, NS) or with aromatase activity in NCM (r S 0.429, NS). based on the fact that AR, and not estrogen receptors alpha (ER ), are colocalized with aromatase in NCM (Metzdorf et al., 1999; Fusani et al., 2000). Estrogen receptors beta (ER ) are expressed at low levels in NCM (Bernard et al., 1999), but at present it is not known whether estrogen can regulate aromatase by binding to ER. It has been reported that estradiol down-regulates aromatase expression in primary cultures of developing zebra finch telencephalon (Freking et al., 1998). However, recent work suggests that these effects are mainly limited to aromatase-expressing hippocampal neurons (Saldanha et al., 2000b). The tissue-specific regulation of aromatase is most likely due to the existence of tissue-specific promoters (Simpson et al., 1994; Lephart, 1996). The telencephalon of the zebra finch expresses predominantly a form of aromatase-mrna containing the exon Ia, which is regulated through an alternate promoter (Ramachandran et al., 1999). At present, we do not know whether this promoter contains an androgen response element that would be required for AR-mediated androgen regulation. Further, double-labeling studies for AR and aromatase are necessary to ascertain whether these two factors are expressed in the same NCM cells. An alternative explanation is that T (or E2 derived from T in the brain) regulates aromatase indirectly by acting trans-synaptically, for example by acting on catecholaminergic neurons that project to NCM. In the Japanese quail, catecholamines regulate aromatase activity in the preoptic area, and this action appears to be mediated by regulation of cyclic AMP (camp) synthesis (Balthazart and Ball, 1998). In primary cultures of developing zebra finch telencephalon, camp up-regulates aromatase activity (Freking et al., 1998). However, there are no known catecolaminergic inputs DISCUSSION This study shows that in female canaries a 4-week T treatment induces an increase in the mrna expression and activity of aromatase in the NCM. In T- treated females, aromatase-mrna expression in the caudal region of NCM was positively correlated with the plasma levels of T. To our knowledge, this is the first direct evidence for a steroid-dependent regulation of aromatase expression and activity in the songbird telencephalon. It is likely that T regulates aromatase expression in NCM by binding to androgen receptors (AR) in the same cells expressing aromatase. We advanced this hypothesis in a previous study (Metzdorf et al., 1999) Figure 5 Expression level of aromatase-mrna (mean S.E.M.) in the rostral and caudal region of the caudomedial neostriatum (NCM) in untreated female canaries and in females implanted for 4 weeks with a T-releasing pellet (25 g/day; T-treated). *p.05, t test.

6 6 Fusani et al. to NCM, and immunocytochemical localization of tyrosine hydroxylase, which is the limiting enzyme in catecholamine synthesis, showed little or no tyrosine hydroxylase immunoreactive fibers in the zebra finch NCM (Bottjer, 1993). Nevertheless, other neurotransmitters could regulate camp, and in turn aromatase, in NCM. The most caudomedial region of HVC [called para-hvc by some authors (Bottjer and Johnson, 1997)], which contains high concentrations of ER (Gahr et al., 1993; Metzdorf et al., 1999), projects to a region of NCM (Foster and Bottjer, 1998) that has the highest concentration of aromatase (Shen et al., 1995; Metzdorf et al., 1999; Saldanha et al., 2000b). Thus, estrogen could act on estrogensensitive neurons in HVC and regulate trans-synaptically aromatase in NCM. It will be important to identify the type(s) of neurotransmitter of these neurons and test whether they can regulate NCM aromatase. At present, it is difficult to speculate on whether the results of our study can be generalized to other songbird species. Few studies have investigated directly the regulation of aromatase in NCM by means of hormone manipulation. In both male and female zebra finches, gonadectomy does not affect aromatase activity in NCM, although T-treatment of gonadectomized birds induces a numerical increase of aromatase activity (Vockel et al., 1990b). Similarly, in male zebra finches castration and castration followed by estradiol replacement do not affect the density of aromataselike immunoreactive somata and fibers in NCM (Saldanha et al., 2000b). Thus, the mechanisms of regulation of aromatase in the zebra finch NCM could differ from those of canaries. For example, adult zebra finches have, compared to canaries, very low expression of ER in HVC (Gahr et al., 1993). In addition, there is correlative evidence for species differences in the androgen-dependent regulation of NCM aromatase. Aromatase activity and/or expression in NCM is seasonally correlated with plasma levels of androgen in the canary (Fusani et al., 2000) and in the Lapland longspur (Soma et al., 1999), but not in the pied flycatcher (Foidart et al., 1998; Silverin et al., 2000), chaffinch, willow warbler, or great tit (Silverin et al., 2000). It is likely that the modalities of regulation of aromatase in NCM can be better understood if investigations take into account the probable functional role of the enzyme in this area. If aromatase provides estrogen for song-related memory functions in NCM, it can be expected that the enzyme is particularly active in periods of song learning and/or plasticity. For example, in canaries, in late autumn and early spring, when song reorganization occurs, T promotes the incorporation of new syllables in the song and of newborn neurons in HVC, and T action appears to be mediated by its conversion into estradiol (Alvarez- Buylla and Kirn, 1997). Therefore, T might increase aromatase activity in periods of song plasticity and thus provide estrogen that facilitates motor or perceptual reorganization of song. In female canaries, an increase in estrogen production consequent to the increase in aromatase concentration that follows the T treatment might be functional in masculinization of the song and the song system. Previous work has shown that the masculinization of song and the song system of female canaries is, at least in part, estrogendependent (DeVoogd and Nottebohm, 1981; De- Voogd, 1986). However, the development of malelike song and the masculinization of the gross morphology of HVC occur also when the aromatase inhibitor Fadrozole is given simultaneously with T to prevent its conversion into estrogen (Fusani, 2000). Nevertheless, estrogen has other effects on the song system and song. The repetition rate of the syllables, which is a critical cue for female preference, is estrogen dependent (Fusani, 2000). T increases the survival of newborn neurons in HVC by increasing the expression of the brain-derived neurotrophic factor (BDNF) (Rasika et al., 1999), and these effects are probably mediated by the aromatization of T into estradiol (Dittrich et al., 1999). Therefore, the increase of aromatase expression and activity in NCM could play a primary role in the dramatic changes induced by T on song and the song system of female canaries. This interpretation requires that the E2 produced in NCM reaches the song system and in particular HVC. At present, there is no evidence that NCM sends projections to HVC, and no direct demonstration that E2 is transported from NCM to HVC. Nevertheless, E2 could reach HVC via microcapillary systems and/or passive transport, that is, diffusion. This possibility is supported by the fact that E2 produced in the brain leaves the areas of synthesis and can be detected in the circulation (Schlinger and Arnold, 1992, 1993). We are grateful to Dr. T. Van t Hof for making available to us his laboratory for the hormone measurements and for helping with the RIA methods. We thank Dr. B. A. Schlinger for his comments on a previous version of the manuscript. REFERENCES Adkins-Regan E, Abdelnabi M, Mobarak M, Ottinger MA Sex steroid levels in developing and adult male and female zebra finches (Poephila guttata). Gen Comp Endocrinol 78:

7 Aromatase Regulation in the Canary Neostriatum 7 Alvarez-Buylla A, Kirn JR Birth, migration, incorporation, and death of vocal control neurons in adult songbirds. J Neurobiol 33: Balthazart J, Ball GF New insights into the regulation and function of brain estrogen synthase (aromatase). Trends Neurosci 21: Balthazart J, Stoop R, Foidart A, Harada H Synergistic control by androgens and estrogens of aromatase in the quail brain. NeuroReport 5: Bernard DJ, Bentley GE, Balthazart J, Turek FW, Ball GF Androgen receptor, estrogen receptor alpha, and estrogen receptor beta show distinct patterns of expression in forebrain song control nuclei of European starlings. Endocrinology 140: Bottjer SW The distribution of tyrosine hydroxylase immunoreactivity in the brains of male and female zebra finches. J Neurobiol 24: Bottjer SW, Johnson F Circuits, hormones, and learning: Vocal behavior in songbirds. J Neurobiol 33: DeVoogd T, Nottebohm F Gonadal hormones induce dendritic growth in the adult avian brain. Science 214: DeVoogd TJ Steroid interactions with structure and function of avian song control regions. J Neurobiol 17: Dittrich F, Feng Y, Metzdorf R, Gahr M Estrogeninducible, sex-specific expression of brain-derived neurotrophic factor mrna in a forebrain song control nucleus of the juvenile zebra finch. Proc Natl Acad Sci USA 96: Foidart A, Silverin B, Baillien M, Harada N, Balthazart J Neuroanatomical distribution and variations across the reproductive cycle of aromatase activity and aromatase-immunoreactive cells in the pied flycatcher (Ficedula hypoleuca). Horm Behav 33: Foster EF, Bottjer SW Axonal connections of the High Vocal Center and surrounding cortical regions in juvenile and adult male zebra finches. J Comp Neurol 397: Freking F, Ramachandran B, Schlinger BA Regulation of aromatase, 5 alpha- and 5 beta-reductase in primary cell cultures of developing zebra finch telencephalon. J Neurobiol 36: Fusani L Action of oestrogen on brain mechanisms of behaviour in the ring dove (Streptopelia risoria) and canary (Serinus canaria). PhD Thesis. Cambridge: University of Cambridge. Fusani L, Van t Hof T, Hutchison JB, Gahr M Seasonal expression of androgen receptors, estrogen receptors and aromatase in the canary brain in relation to circulating androgens and estrogens. J Neurobiol 43: Gahr M, Güttinger H-R, Kroodsma DE Estrogen receptors in the avian brain: Survey reveals general distribution and forebrain areas unique to songbirds. J Comp Neurol 327: Gahr M, Metzdorf R Distribution and dynamics in the expression of androgen and estrogen receptors in vocal control systems of songbirds. Brain Res Bull 44: Güttinger H-R The integration of learnt and genetically programmed behaviour: A study of hierarchical organization in songs of canaries, greenfinches and their hybrids. Z Tierpsychol 49: Heid P, Güttinger H-R, Pröve E The influence of castration and testosterone replacement on the song architecture of canaries (Serinus canaria). Z Tierpsychol 69: Hutchison JB, Steimer T Androgen metabolism in the brain: Behavioural correlates. In: de Vries GJ, de Bruin JPC, Uylings HBM, Corner MA, editors. Sex Differences in the Brain. Amsterdam: Elsevier, p Lephart ED A review of brain aromatase cytochrome P450. Brain Res Rev 22:1 26. Margoliash D Functional organization of forebrain pathways for the song production and perception. J Neurobiol 33: Marler P, Peters S, Wingfield J Correlations between song acquisition, song production, and plasma levels of testosterone and estradiol in sparrows. J Neurobiol 18: Metzdorf R, Gahr M, Fusani L Distribution of aromatase, estrogen receptor, and androgen receptor mrna in the forebrain of songbirds and nonsongbirds. J Comp Neurol 405:1 15. Nottebohm F, Nottebohm ME Relationship between song repertoire and age in the canary, Serinus canaria. Z Tierpsychol 46: Ramachandran B, Schlinger BA, Arnold AP, Campagnoni AT Zebra finch aromatase gene expression is regulated in the brain through an alternate promoter. Gene 240: Rasika S, Alvarez-Buylla A, Nottebohm F BDNF mediates the effects of testosterone on the survival of new neurons in an adult brain. Neuron 22: Roselli CE, Resko JA Androgens regulate brain aromatase activity in adult male rats through a receptor mechanism. Endocrinology 114: Saldanha CJ, Schlinger BA Estrogen synthesis and secretion in the brown-headed cowbird (Molothrus ater). Gen Comp Endocrinol 105: Saldanha CJ, Schultz JD, London SE, Schlinger BA. 2000a. Telencephalic aromatase but not a song circuit in a suboscine passerine, the golden-collared manakin (Manacus vitellinus). Brain Behav Evol 56: Saldanha CJ, Tuerk MJ, Kim YH, Fernandes AO, Arnold AP, Schlinger BA. 2000b. Distribution and regulation of telencephalic aromatase expression in the zebra finch revealed with a specific antibody. J Comp Neurol 423: Schlinger BA The activity and expression of aromatase in songbirds. Brain Res Bull 44: Schlinger BA, Arnold AP Circulating estrogens in a male songbird originate in the brain. Proc Natl Acad Sci USA 89:

8 8 Fusani et al. Schlinger BA, Arnold AP Estrogen synthesis in vivo in the adult zebra finch: additional evidence that circulating estrogens can originate in the brain. Endocrinology 133: Schlinger BA, Callard GV Aromatase in quail brain: A functional relationship to aromatase and aggressiveness. Endocrinology 124: Schumacher M, Balthazart J Neuroanatomical distribution of testosterone-metabolizing enzymes in the Japanese quail. Brain Res 422: Shen P, Schlinger BA, Campagnoni AT, Arnold AP An atlas of aromatase mrna expression in the zebra finch brain. J Comp Neurol 360: Silverin B, Baillien M, Foidart A, Balthazart J Distribution of aromatase activity in the brain and peripheral tissues of passerine and nonpasserine avian species. Gen Comp Endocrinol 117: Simpson ER, Mahendroo MS, Means GD, Kilgore MW, Hinselwood MM, Graham-Lorence S, Amarneh B, Ito Y, Fisher CR, Michael MD Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis. Endocrine Rev 15: Soma KK, Bindra RK, Gee J, Wingfield JC, Schlinger BA Androgen-metabolizing enzymes show region-specific changes across the breeding season in the brain of a wild songbird. J Neurobiol 41: Steimer T, Hutchison JB Androgen increases formation of behaviourally effective oestrogen in the dove brain. Nature 292: Steimer T, Hutchison JB Is the androgen-dependent increase in preoptic estradiol-17 formation due to aromatase induction? Brain Res 480: Vockel A, Pröve E, Balthazart J. 1990a. Sex- and agerelated differences in the activity of testosterone-metabolizing enzymes in microdissected nuclei of the zebra finch brain. Brain Res 511: Vockel A, Pröve E, Balthazart J. 1990b. Effects of castration and testosterone treatment on the activity of testosterone-metabolizing enzymes in the brain of male and female zebra finches. J Neurobiol 21: Weichel K, Schwager G, Heid P, Güttinger H-R, Pesch A Sex differences in plasma steroid concentrations and singing behaviour during ontogeny in canaries (Serinus canaria). Ethology 73: Whitfield HJ, Brady LS, Smith MA, Mamalaki E, Fox RJ, Herkenham M Optimization of crna probe in situ hybridization methodology for localization of glucocorticoid receptor mrna in rat brain: a detailed protocol. Cell Mol Neurobiol 10: