Estradiol modulates brainstem catecholaminergic cell groups and projections to the auditory forebrain in a female songbird

BR A I N R ES E A RC H 1 1 7 1 ( 2 00 7 ) 9 3 –1 03

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s

Research Report

Estradiol modulates brainstem catecholaminergic cell groups and projections to the auditory forebrain in a female songbird Meredith M. LeBlanc a , Christopher T. Goode a , Elizabeth A. MacDougall-Shackleton b , Donna L. Maney a,⁎ a

Department of Psychology, 532 Kilgo Circle, Emory University, Atlanta, GA 30322, USA Department of Biology, University of Western Ontario, London, Ontario, Canada

b

A R T I C LE I N FO

AB S T R A C T

Article history:

In songbirds, hearing conspecific song induces robust expression of the immediate early

Accepted 19 June 2007

gene zenk in the auditory forebrain. This genomic response to song is well characterized in

Available online 27 July 2007

males and females of many species, and is highly selective for behaviorally relevant song. In

Keywords:

estradiol; we previously showed that in non-breeding females with low levels of plasma

Estrogen

estradiol, the zenk response to hearing song is no different than the response to hearing

Auditory

frequency-matched tones. Here, we investigated the role of brainstem catecholaminergic

Norepinephrine

cells groups, which project to the forebrain, in estradiol-dependent selectivity. First, we

Dopamine

hypothesized that estradiol treatment affects catecholaminergic innervation of the auditory

Song

forebrain as well as its possible sources in the brainstem. Immunohistochemical staining of

Mate choice

tyrosine hydroxylase revealed that estradiol treatment significantly increased the density of

white-throated sparrows, the selectivity of the zenk response requires breeding levels of

catecholaminergic innervation of the auditory forebrain as well as the number of catecholaminergic cells in the locus coeruleus (A6) and the ventral tegmental area (A10), both of which are known to contain estrogen receptors in songbirds. Second, we hypothesized that during song perception, catecholaminergic cell groups of the brainstem actively participate in auditory selectivity via estrogen-dependent changes in activity. We found that hearing songs did not induce the expression of zenk, a putative marker of activity, within catecholaminergic neurons in any of the cell groups quantified. Together, our results suggest that estradiol induces changes in brainstem catecholaminergic cell groups that may play a neuromodulatory role in behavioral and auditory selectivity. © 2007 Elsevier B.V. All rights reserved.

1.

Introduction

Lengthening days in the early spring mark the beginning of a chain of events that affect hormones and behavior in many species. Even a small change in photoperiod can trigger shifts in the hormonal state of songbirds such as white-throated sparrows (Zonotrichia albicollis), causing their gonads to grow

(Wolfson, 1958; Shank, 1959). The resulting marked change in gonadal steroids during this time supports the development of seasonally appropriate reproductive behavior. For example, in female songbirds, behavioral responses to male song change dramatically when estradiol (E2) levels rise. During the breeding season, females of many species respond to audio recordings of male song with a stereotyped behavior known as

⁎ Corresponding author. Fax: +1 404 727 0372. E-mail address: [email protected] (D.L. Maney). 0006-8993/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2007.06.086

94

BR A IN RE S E A RCH 1 1 71 ( 20 0 7 ) 9 3 –1 03

copulation solicitation display (CSD); during the non-breeding season, however, females do not perform CSDs even when presented with song that in the breeding season would be highly stimulatory (Kern and King, 1972; Moore, 1983). This robust change in behavioral responses to sociosexual auditory stimuli suggests that E2 may act within the auditory system to affect the processing of auditory signals. Song-induced auditory responses in the brain are often studied by quantifying the expression of the immediate early gene zenk (zif-268, egr-1, ngfi-a, and krox24; Mello et al., 1992) or its protein product (ZENK; Mello and Ribeiro, 1998). This “genomic response” to song is well-characterized in males and females of many species, including zebra finches (Taeniopygia guttata; Bailey et al., 2002; Mello and Clayton, 1994; Mello et al., 1992; Stripling et al., 2001), canaries (Serinus canaria; Leitner et al., 2005; Ribeiro et al., 1998; Terleph et al., 2006), and European starlings (Sturnus vulgaris; Gentner et al., 2001; Sockman et al., 2002). The magnitude of this response in the caudomedial nidopallium (NCM), a region of the auditory forebrain, is greater in response to song than to synthetic tones, and greater to conspecific than heterospecific song (Mello and Clayton, 1994; Stripling et al., 2001). Social factors known to affect the magnitude of behavioral responses, such as song complexity, dialect, or familiarity to the listener, also affect the genomic response (Gentner et al., 2001; Leitner et al., 2005; Maney et al., 2003; Sockman et al., 2002; Terpstra et al., 2006). Thus, the magnitude of this response relates to the behavioral relevance of the stimulus. Just as E2 appears to alter the behavioral relevance of song, it also affects song-induced genomic responses in the auditory forebrain. We previously showed that the selectivity of the genomic response in NCM requires breeding levels of E2. In non-breeding female white-throated sparrows with low levels of plasma E2, the genomic response to hearing songs is not distinguishable from the response to hearing frequencymatched tones (Maney et al., 2006). The plastic nature of auditory selectivity suggests that E2 modulates auditory pathways and processing centers to promote recognition of and attention to conspecific song during the breeding season. Increased auditory selectivity could be related to one or more cognitive processes that depend on catecholamines (CAs). CAs, particularly norepinephrine from the locus coeruleus, are widely known to shape the response properties of sensory networks to alter selectivity (see Hurley et al., 2004 for review) and may play a role in selective attention (see AstonJones and Cohen, 2005, for review). In songbirds, CA projections to the forebrain have been hypothesized to affect the auditory processing of as well as behavioral responses to behaviorally relevant social signals such as song (e.g., Appeltants et al., 2002a,b; 2005; Bharati and Goodson, 2006; Cardin and Schmidt, 2004; Maney and Ball, 2003; Riters and Pawlisch, 2007). In female canaries, noradrenergic denervation of the forebrain causes a reduction in CSD behavior along with an apparent deficit in selective attention to sexually stimulating song (Appeltants et al., 2002b). If CA fibers and their sources are sensitive to gonadal steroids, they may mediate seasonal changes in selective attention and auditory responses to these signals. A large literature indicates that in mammals, CA neurons are in fact targets of E2. E2 treatment increases mRNA for tyrosine hydroxylase (TH) and dopamine beta-hydroxylase

(DBH), rate-limiting enzymes in the synthesis of CAs, in brainstem CA cell groups (Pau et al., 2000; Serova et al., 2002, 2004). E2 may also affect developing CA cells; E2 treatment promotes the expression of TH mRNA as well as neurite branching in cultured embryonic midbrain cells (Ivanova and Beyer, 2003; Kuppers et al., 2000). Kritzer and Kohama (1998, 1999) reported that in rhesus monkeys, TH and DBH immunoreactivity (IR) in the forebrain is depleted by ovariectomy and restored by ovarian hormone replacement. In songbirds, forebrain CA turnover as well as adrenergic receptor density are modulated seasonally by gonadal steroids (Barclay and Harding, 1990; Riters et al., 2002). CAs therefore represent an excellent candidate system for mediating seasonal changes in auditory and behavioral responses to sociosexual stimuli. In the present study, we looked for evidence that the E2induced plasticity in the auditory forebrain described by Maney et al. (2006) is mediated by CAs. Our first hypothesis was that E2 alters the CA innervation of the auditory forebrain as well as the possible sources of this innervation in the brainstem. To test this hypothesis, we measured the effects of E2 treatment on the density of TH-IR innervation of NCM as well as the number of TH-IR cells in brainstem CA cell groups (Fig. 1). Although the exact origin of CA fibers in NCM (Fig. 2) is currently unknown, tract tracing studies have demonstrated that CA cell groups A6, A9, A10, and A11 project to areas of the canary forebrain involved in song learning and production (Appeltants et al., 2000, 2002a). Thus, it is plausible that CA innervation of NCM originates in one or more of these regions. We predicted E2-dependent changes in TH-IR both in NCM and in these CA brainstem cell groups. Our second hypothesis was that these CA cells directly regulate forebrain selectivity by altering their activity, and therefore CA synaptic activity in the forebrain, during song perception. To test this hypothesis, we quantified transcription activity, a putative measure of depolarization activity (see Mello et al., 2004), by counting the number of TH-IR cells that were immunopositive for ZENK protein in each CA cell group of the brainstem after the birds listened to song or tone stimuli. We predicted that the zenk response in CA cells would parallel and therefore possibly contribute toward E2-dependent selectivity in NCM (Maney et al., 2006).

2.

Results

2.1.

Behavioral analysis

Only E2-treated birds who heard songs performed CSDs during the stimulus presentation, which confirmed that the implants raised plasma E2 levels and that the tones were not interpreted as songs (Maney et al., 2006). Within this group (n = 6), there were no significant correlations between CSD behavior and any of the neural variables we quantified, demonstrating that the variation in these variables was not completely explained by variation in CSD behavior.

2.2.

Effects of E2 treatment on TH immunoreactivity

A MANOVA revealed a significant effect of treatment on TH immunoreactivity in our regions of interest (F1,21 = 3.042,

BR A I N R ES E A RC H 1 1 7 1 ( 2 00 7 ) 9 3 –1 03

95

Fig. 4) or by type of stimulus (F1,21 = 0.498, p = 0.738). There was no interaction between E2 treatment and type of stimulus on the number of double-labeled ZENK-TH cells in these areas (F1,21 = 0.400, p = 0.806). There was no colocalization of TH-IR and ZENK-IR in A6 (see Fig. 5A and Section 2.5).

2.4.

Assessment of plasma E2 levels

Plasma E2 levels fell within a physiological range in the E2treated birds (1.3 ± 0.2 ng/ml) and were significantly higher than in the blank-implanted birds ( p = 0.007). The ovaries of all birds in the blank-implanted group were completely regressed with no visible follicles.

Fig. 1 – CA immunoreactivity was quantified in six regions of interest: (A) midbrain central gray (A11) and ventral tegmental area (A10); (B) caudomedial nidopallium (NCM) and substantia nigra (A9); (C) rostral locus coeruleus (A8); and (D) caudal locus coeruleus (A6). The robust nucleus of the arcopallium (RA) is shown as a landmark that was used to locate NCM.

p = 0.044; Figs. 2 and 3). Post-hoc LSD tests revealed the effect was significant in NCM, measured either by the percent area stained ( p = 0.047) or by the optical density ( p = 0.027). E2 also increased the number of TH-IR cells in A6 ( p = 0.018) and A10 ( p = 0.004) but not in A8 ( p = 0.834), A9 ( p = 0.876), or A11 ( p = 0.260).

2.3. Effect of E2 treatment and type of stimulus on ZENK induction in TH-IR cells The number of double-labeled ZENK-TH cells in brainstem cell groups was not affected by E2 treatment (F1,21 = 0.662, p = 0.627;

Fig. 2 – TH-IR fibers in NCM (see Fig. 1B for location). In birds treated with E2 (A), TH-IR fibers covered a larger percentage of the area sampled than in birds treated with blank implants (B). See also Fig. 3. Staining in panels A and B above represents levels closest to the mean for each group. The shapes that resemble cell bodies, such as the one indicated by the arrow, are actually basket-like structures formed by the fibers as they surround TH-immunonegative cells (see also Appeltants et al., 2001). There are no TH-IR cell bodies in NCM.

96

BR A IN RE S E A RCH 1 1 71 ( 20 0 7 ) 9 3 –1 03

Fig. 3 – Effects of E2 treatment on the percent area covered by TH-IR fibers in NCM and the number of TH-IR cells in CA cell groups A6, A8, A9, A10, and A11. *p b 0.05; **p b 0.02; ***p b 0.005.

Fig. 4 – ZENK-IR induced in CA cell groups A8, A9, A10, and A11 by songs or tones in birds treated with E2 or blank implants. Note the low numbers of double-labeled cells. There were no double-labeled cells in A6.

BR A I N R ES E A RC H 1 1 7 1 ( 2 00 7 ) 9 3 –1 03

97

birds. These results confirm that we can detect ZENK-IR in CA A6 cells.

3.

Discussion

The present study is an extension of work by Maney et al. (2006), who reported that in female white-throated sparrows, E2 treatment increased selectivity of the zenk response in the auditory forebrain. In that study, the zenk response was selective for songs vs. synthetic tones only in females with breeding-like levels of E2. A possible explanation for a sharpening of the zenk response is that E2 treatment affects CA innervation of that region. We report here that E2 treatment increased the density of CA fibers in NCM, a region of the auditory forebrain (Figs. 2 and 3), as well as in two CA cell groups of the brainstem, A6 and A10 (Fig. 3). E2 treatment did not, however, affect the genomic responses of CA cells to auditory stimuli (Fig. 4), suggesting that these CA neurons may not produce direct excitatory or inhibitory effects on the auditory forebrain during stimulus processing but rather may play a neuromodulatory role in E2-dependent selectivity.

3.1.

Fig. 5 – TH- and ZENK-immunopositive neurons in this study. (A) Single-labeled TH-IR neurons (e.g., white arrow) and ZENK-IR nuclei (e.g., black arrow) in cell group A6. The TH immunoreactivity is localized in the cell bodies and fibers, whereas the ZENK immunoreactivity is localized in the cell nucleus. (B) Double-labeled neurons (e.g., black arrow) in cell group A11. Each of the TH-IR neurons, indicated by the brown cytoplasmic stain, is also immunopositive for ZENK, as indicated by the stained nucleus. Single-labeled ZENK cells are also visible.

2.5.

Induction of ZENK-IR in A6 cells

In order to evaluate our ability to detect ZENK-IR in the A6 population, we injected a separate set of females with Nmethyl-D,L-aspartic acid (NMDA) or saline and looked for colocalization of ZENK protein and dopamine beta-hydroxylase (DBH), an enzyme involved in norepinephrine synthesis, in A6. In the NMDA-treated birds, nearly every DBH-IR cell in A6 was also immunopositive for ZENK. Some double-labeled ZENK-DBH cells were also seen in A6 of the saline-treated

E2 modulates CA fiber density in the auditory forebrain

E2 treatment significantly increased the density of CA innervation of NCM (Fig. 2), whether measured by the percent area covered by TH-IR fibers (Fig. 3) or by the optical density of the area sampled. This result is consistent with the hypothesis that CA innervation of sensory processing areas plays a role in sharpening neural responses to sensory stimuli, particularly signals important for mate choice. In mammals, CAs are hypothesized to act as endogenous neuromodulators that filter stimuli and tune sensory systems, allowing them to respond selectively to stimuli most relevant to survival and reproduction (reviewed by AstonJones and Cohen, 2005; Hurley et al., 2004). CA systems are targets of E2 and are modulated by hormonal manipulations in rodents, primates, and songbirds (Barclay and Harding, 1990; Ivanova and Beyer, 2003; Kritzer and Kohama, 1998, 1999; Kuppers et al., 2000; Riters et al., 2002; Serova et al., 2002, 2004). Our findings extend those by showing that CA innervation of an auditory processing area is modulated by hormones and may explain hormone-dependent changes in auditory selectivity. Because TH is involved in the synthesis of both dopamine and norepinephrine, our anti-TH antibody does not distinguish between these two transmitters. It is likely that both are found in NCM. Dopaminergic and noradrenergic fibers are widespread throughout the telencephalon in birds (Reiner et al., 1994). Immunolabeling of TH reveals more fibers than does immunolabeling of DBH, which is present in norepinephrine but not dopamine cells and fibers, suggesting that many TH-IR fibers represent dopamine processes. Barclay and Harding (1988, 1990) reported the presence of both dopamine and norepinephrine at many locations throughout the forebrain in zebra finches. Our own work has revealed fibers immunopositive for DBH in NCM in this species (Maney, unpublished observations), which is consistent with a study in zebra

98

BR A IN RE S E A RCH 1 1 71 ( 20 0 7 ) 9 3 –1 03

finches (Mello et al., 1998). Future studies will investigate the effects of gonadal steroids on DBH innervation of the auditory forebrain in both males and females.

3.2.

E2 modulates CA cell number in A10 and A6

We counted more TH-IR neurons in CA cell groups A6 and A10 in the E2-treated birds than in birds with blank implants. This result is consistent with evidence in other species that estrogen increases the synthesis of TH and DBH (Ivanova and Beyer, 2003; Serova et al., 2002, 2004). A previous study of the distribution of ER-alpha mRNA in brainstem CA cell groups of the canary showed that among our regions of interest, A6 and A10 contain ER-alpha whereas the other cell groups do not (Maney et al., 2001). It is possible, therefore, that E2 affects TH synthesis by acting directly on these cells. In that study, ER-alpha was found to be expressed in the rostral portion of A10, whereas the E2-induced changes in CA cell number reported in the present study were found more caudally. A6, on the other hand, is a relatively small area; neither the distribution of ER-alpha nor the effects of E2 on TH-IR cell number were localizable to a distinct subregion within it. Maney et al. (2001) did not perform double-labeling to see whether ER-alpha was expressed within TH-IR neurons; however, evidence from the mammalian literature suggests that in both A6 and A10, TH and ER-alpha are colocalized in the same cells (e.g., Helena et al., 2006; Kritzer, 1997). ERalpha is also found in NCM in both male and female songbirds of many species (e.g., Gahr et al., 1993); E2 may thus modulate auditory processes by acting directly on NCM as well as on receptors in CA cell groups. Do any of the CA cell groups in the brainstem project to the auditory forebrain? The tract tracing studies conducted in songbirds have focused primarily on projections to the song system rather than to auditory regions, and in males rather than females. Injections of retrograde tracer into song control nuclei, some of which are located near and are interconnected with NCM, reveal inputs from A6, A9, A10 and A11 (Appeltants et al., 2000, 2002a, Castelino et al., 2007). In birds in general, heavy projections of the midbrain dopaminergic cell groups, particularly A9 and A10, into pallial forebrain regions are welldescribed (Reiner et al., 1994) and are more likely sources of NCM innervation than the hypothalamic cell groups such as A14 and A15, which most likely project locally or caudally rather than to more rostral telencephalic areas (Cabot et al., 1982). DBH-IR fibers in NCM (Maney, unpublished observations; Mello and Ribeiro, 1998) must arise from forebrainprojecting noradrenergic cell groups, which are limited primarily to A6 (reviewed by Reiner et al., 1994; see also Castelino et al., 2007). We are confident, therefore, that despite the lack of definitive tract-tracing studies, our regions of interest are well chosen. Future work will directly investigate the sources of TH- and DBH-IR innervation of NCM.

3.3.

Direct vs. indirect role of CA cells in auditory selectivity

The effect of E2 on forebrain auditory selectivity reported by Maney et al. (2006) was not attributable to E2-dependent increases in the genomic response to song. Rather, E2 affected selectivity by reducing the response to frequency-matched

tones, which elicited lower numbers of ZENK-immunopositive cells in E2-treated birds than in the other groups. This pattern suggests E2-dependent modulatory input to the auditory forebrain that sharpens responses to behaviorally relevant signals by inhibiting the response to non-relevant stimuli. Such filtering has been proposed to involve CA systems, notably the A6 and A10 cell groups and their projections to the forebrain (reviewed by Aston-Jones and Cohen, 2005). In rats and monkeys, salient stimuli such as tapping the cage door caused activation of A6 neurons along with a behavioral orienting response (e.g., Foote et al., 1980; Grant et al., 1988). Electrical stimulation of the A10 cell group, when paired with 9 kHz tones, caused remodeling of the auditory cortex in rats so that 9 kHz is overrepresented—suggesting that A10 activation alters the perceived salience of that stimulus (Bao et al., 2001). Here, we hypothesized that CA cell groups of the brainstem participate directly in the modulation of auditory responses to song by changing their activity during song processing. We predicted that songs and tones would induce similar zenk responses in CA neurons in females treated with blank implants, and that in the E2-treated birds, songs and tones would induce significantly different zenk responses. A song-induced response that was either higher or lower than a tone-induced response in E2-treated birds would have supported our hypothesis. We found, however, no effects of either hormone treatment or type of stimulus on ZENK-IR in CA cells, and no interactions between the two factors. The numbers of TH-IR cells that expressed ZENK-IR were extremely low (Fig. 4), averaging fewer than ten and in some regions fewer than two double-labeled cells per bird. These low numbers are similar to those reported by Bharati and Goodson (2006), who quantified immediate early gene expression in dopaminergic cell groups in zebra finches responding to various sociosexual stimuli. When taken together with their study, our results suggest that dopaminergic cells in A9–A11 do not undergo changes in transcription activity, as would be indicated by immediate early gene expression, in response to behaviorally relevant auditory stimuli. Whether these cells change their firing rates was not directly measured here, but immediate early genes such as zenk are widely recognized as activity-dependent and increase transiently in response to stimuli that lead to depolarization (see Mello et al., 2004 for review). Some populations of neurons do not express ZENK even when they are presumably undergoing significant depolarization; for example, ZENK-IR is notably absent in Field L2 of the auditory forebrain (Mello and Ribeiro, 1998) and in nucleus rotundus of the thalamus (Maney, unpublished observations). Because we saw no colocalization of ZENK-IR and TH-IR in the A6 population, we experimentally evaluated our ability to detect ZENK-IR in these cells by inducing its expression with injections of NMDA (Saitoh et al., 1991). This treatment induced massive zenk activation of DBH-IR cells in the A6 cell group, validating our immunolabeling procedure and confirming that this cell group can and does express ZENK-IR when stimulated by a glutamate agonist. These cells did not, however, show a zenk response to auditory stimuli in our study. It is therefore feasible that cells in A6, as well as those in our other regions of interest, are not themselves selectively activated by behaviorally relevant song stimuli.

BR A I N R ES E A RC H 1 1 7 1 ( 2 00 7 ) 9 3 –1 03

The relative genomic quiescence of the brainstem CA nuclei during song processing suggests a neuromodulatory rather than directly excitatory or inhibitory role for these regions in E2-dependent auditory selectivity. Some CA terminals may release transmitter in a non-synaptic, or paracrine fashion, which may alter the responsivity or spontaneous firing activity in forebrain neurons (reviewed by Beaudet and Descarries, 1978). Such changes, which may be sustained over prolonged periods (Reader et al., 1979) may enable forebrain areas to respond differently to the same input depending on the behavioral state of the animal (e.g., season or social context), without stimulus-induced changes in the CA neurons themselves. In male zebra finches, depletion of forebrain norepinephrine abolishes the effect of social context on zenk expression in a forebrain song area (Castelino and Ball, 2005). We propose that seasonal changes in CA neuromodulation enable auditory systems of female songbirds to respond selectively to male song during the breeding season, and to respond less selectively outside it. Most contemporary models of CA attentional filtering do take into account selective activation of CA neurons, however (Aston-Jones and Cohen, 2005), and the activity of CA cell groups during song processing should be investigated using electrophysiological techniques.

3.4.

Role of CAs in CSD behavior

In order to confirm that variation in TH-IR fiber density, numbers of TH-IR cells, and double-labeled TH-ZENK cells was not due entirely to variation in CSD behavior, we looked for correlations between these neural variables and CSD behavior and found none. This result should not, however, be taken as evidence against a role for CAs in behavioral responses or in the encoding of behavioral salience. The number of birds performing CSDs in response to auditory stimuli in this study was only five, which, although it allows us to demonstrate that our neural variables were not completely explained by CSD behavior, is far too few to properly test whether CAs might play a role in this behavior or other responses to song. Others have tested this hypothesis more directly (Appeltants et al., 2002b; Riters and Pawlisch, 2007). For example, Appeltants et al. (2002b) treated female canaries with a neurotoxin that depletes the forebrain, including NCM, of noradrenergic fibers. They then demonstrated that treated females performed fewer CSDs in response to sexually stimulating songs presented either alone or with auditory distracters. The treatment did not affect other estrogen-dependent behaviors or locomotor activity, suggesting that the effect on CSD behavior was related to auditory perception. In the present study, we show evidence that during the time of year that females are listening to courtship song and responding with CSDs, noradrenergic innervation of the auditory forebrain may reach a peak. In the non-breeding season, when females do not respond to male song by performing CSDs, this innervation may be significantly reduced. In addition, we previously showed evidence that the auditory forebrain is highly selective for male song only during the breeding season (Maney et al., 2006). Our data are therefore consistent with the hypothesis that CAs contribute toward auditory processing of courtship song, and that E2-dependent seasonal changes in the CA system may

99

underlie seasonal rhythms in auditory selectivity and behavioral responsiveness.

4.

Experimental procedure

4.1.

Animals

All procedures involving animals were approved by the Emory University Institutional Animal Care and Use Committee. Twenty-three female white-throated sparrows were collected in mist-nets on the campus of Emory University in Atlanta, GA during November 2004. A small blood sample was taken, and sex was confirmed by PCR analysis using primers P2 and P8 of Griffiths et al. (1998). The birds were housed in the Emory animal care facility in walk-in flight cages and supplied ad libitum with food and water. Day length was kept constant at 10:14 h light–dark, which corresponds to the shortest day they experience while over-wintering at the site of capture. Birds were held under these conditions for approximately 4 months to ensure that photorefractoriness was broken (see Wolfson, 1958; Shank, 1959). No birds were photostimulated in this study and the day length remained at 10:14 h light–dark throughout the experiment.

4.2.

Hormonal manipulation

Before the start of the experiment, birds were moved to individual cages (38 × 38 × 42 cm3) inside identical walk-in sound attenuated booths (Industrial Acoustics, Bronx, NY), each of which held four birds. On the day the birds were moved, each received one subcutaneous silastic implant (length 12 mm, ID 1.47 mm, OD 1.96 mm, Dow Corning, Midland, MI) sealed at both ends with A-100-S Type A medical adhesive (Factor 2, Lakeside, AZ). 11 birds received empty implants and 12 received implants containing 17β-estradiol (Steraloids, Newport, RI). The E2 implants raise and maintain breeding season-like E2 levels for at least 80 days (Moore, 1983). Once they had received implants, birds stayed in the booths an average of 9 days before being exposed to experimental stimuli. Hormone treatment was heterogeneous within housing groups.

4.3.

Stimulus presentation

The auditory stimuli and presentation protocol have been previously described (Maney et al., 2006). Briefly, recordings of singing male white-throated sparrows were downloaded from the Borror Laboratory of Bioacoustics birdsong database and edited so that there was 15 s of silence between songs. The resulting segments were spliced together to form stimulus presentations so that the song of a novel male began every 3 min. Each presentation contained the songs of 14 males, and totaled 42 min in duration. For each of the 14 recordings, the frequency of each whistle (note) in one song was measured and sinusoidal tones were computer-generated at these frequencies and arranged in a random order 200 ms apart. This resulted in tone sequences that matched individual songs in the average number of onsets and offsets as well as total sound energy at each frequency. Tone sequences were spliced

100

BR A IN RE S E A RCH 1 1 71 ( 20 0 7 ) 9 3 –1 03

together in the same manner as the song stimuli, with 15 s of silence between tone sequences. On the afternoon prior to stimulus presentation, each bird's cage was placed in an empty, sound-attenuated booth equipped with microphone, speaker, and video camera. Stimulus playbacks began via the speaker located inside the booth, 1 h after the lights came on the following morning. All stimuli were delivered at a peak level of 70 dB, measured at the bird's cage. Each bird heard either songs (n = 12) or tones (n = 11), followed by 18 min of silence.

4.4.

Behavioral observation

The birds' responses to the auditory stimuli were videotaped, and copulation solicitation display was quantified by counting the number of wing quivers, tail lifts, and trills during the stimulus presentation (Maney et al., 2003, 2006).

4.5.

Histology

Sixty minutes after the start of the stimulus presentation, birds were killed and their brains were processed as described by Maney et al. (2006). A blood sample was retained for E2 radioimmunoassay (see below). Fixed brains were cut coronally on a microtome into 50-μm sections. Every third section caudal to the junction of the diencephalon and midbrain was used in the present study. The dorsal portions of each section, which consisted of telencephalic areas such as NCM, were manually separated from the ventral portions, which contained the brainstem catecholaminergic cell groups. The dorsal and ventral portions were then processed in separate runs of immunocytochemistry (ICC) so that the telencephalon was labeled for TH only, whereas the brainstem was doublelabeled for ZENK and TH.

4.5.1.

Single labeling

TH was labeled in the telencephalon using a standard ICC protocol (Maney et al., 2001, 2003, 2005). Sections were incubated with mouse monoclonal anti-TH antibodies (Immunostar; Hudson, WI) diluted 1:2000 (see Maney et al., 2001). The specificity of this antibody for use in avian species has been verified elsewhere (Bailhache and Balthazart, 1993). TH was subsequently labeled using a biotinylated secondary antibody and the ABC method (Vector, Burlingame, CA) followed by a diaminobenzidine color reaction (Maney et al., 2001) which produces a reddish brown product in the cytoplasm of the cell (Fig. 5).

4.5.2.

Double-labeling

ZENK-IR was visualized in brainstem sections using a previously published protocol (Maney et al., 2003). Briefly, sections were incubated with an antibody against the protein product of zenk (anti-egr-1; Santa Cruz Biotechnology, Santa Cruz, CA), which was subsequently labeled using a biotinylated secondary antibody and the ABC method. Labeling was visualized using diaminobenzidine enhanced with nickel (see Shu et al., 1988; Maney et al., 2003), which results in a bluish black reaction product contained within the cell nucleus (Fig. 5). Sections were then washed and stained for TH-IR as described above for single labeling. After the ICC, all sections

were mounted on slides and coverslipped in DPX as previously described (Maney et al., 2003, 2005, 2006).

4.6.

Quantification of TH-IR in NCM

TH-IR was quantified in NCM on the same coronal plane as the rostralmost area of the robust nucleus of the arcopallium (see Maney et al., 2006; Mello and Clayton, 1994; Bailey et al., 2002). We found previously that in both males and females of this species, gonadal steroids affect the selectivity of the ZENK response maximally in this area of NCM (Maney et al., 2006; LeBlanc, unpublished observations). The sections were photographed using the 10× objective on a Zeiss Axioskop microscope attached to a Leica DFC480 camera. An area within a 0.4 mm2 box placed as close as possible to the midline, adjacent to the hippocampus (Maney et al., 2006), was photographed under a standard light level. All photographs were taken in a single sitting and were approximately 4.4 MB. The images were converted to 8-bit scale, and the thresholding feature of Image J software (NIH) was used to select TH-IR fibers (Maney et al., 2005). The percent area stained was quantified by dividing the area of TH-stained tissue by the total area of NCM sampled. Optical density was quantified by subtracting the optical density of adjacent unstained tissue (usually an area of the hippocampus) from the optical density of the selected area of NCM. Darker staining results in a lower optical density (black = 0, white = 255), so the absolute value of the difference was used in the statistical analysis.

4.7.

Quantification of TH- and ZENK-IR in the brainstem

Five cell groups were examined (Fig. 1): the midbrain central grey (A11), the ventral tegmental area (A10), substantia nigra (A9), rostral locus coeruleus (A8; Reiner et al., 2004), and caudal locus coeruleus (A6; Reiner et al., 1994). These regions were defined using previously established boundaries (Appeltants et al., 2001; Balthazart and Ball, 1996; Bharati and Goodson, 2006; Reiner et al., 1994, 2004; Stokes et al., 1974). Using the 20× objective on the microscope, TH-IR cell bodies were counted in both hemispheres in consecutive sections (150 μm apart) of our regions of interest as previously described (see Fig. 1; also Bharati and Goodson, 2006; Maney and Ball, 2003; Maney et al., 2001). Briefly, for A6, each TH neuron was counted in the entire cell group, which was either two or three sections thick. For A9, the rostral boundary of our quantification was the point at which the cell group assumes a thick, mustache-like shape (Fig. 1B); this area extended caudally for three sections. A8 is continuous with the caudal portion of A9 (Fig. 1C), and in our study continued caudally for two sections. We defined A10 and A11 in a manner consistent with Maney and Ball (2003). The A10 cell group lies lateral to the oculomotor nerve and extends approximately 600 μm (four sections) caudally to the point at which it becomes continuous with the rostral border of A9 (Balthazart and Ball, 1996). For A11, we used three consecutive sections in which this region assumes the shape of a dolphin's tail (Fig. 1A). The number of TH-immunopositive cells that were also immunopositive for ZENK protein (exhibiting a bluish black nucleus within a reddish brown soma; see Fig. 5B) was noted for each region of interest.

BR A I N R ES E A RC H 1 1 7 1 ( 2 00 7 ) 9 3 –1 03

4.8.

Radioimmunoassay for estradiol

Plasma E2 concentration was quantified using a commercial radioimmunoassay kit (Diagnostic Products, Los Angeles, CA) with modifications as described by Pazol et al. (2004). The range of sensitivity was 13.02–2603.90 pg/ml, and the intraand interassay coefficients of variation were 7.95% and 9.56%, respectively.

4.9.

Statistical analyses

The TH-IR data for each region of interest were entered into a single MANOVA with E2 treatment as a between-subjects factor. Post-hoc Fisher's PLSD tests were used to test for effects of E2 within each region of interest. The number of doublelabeled cells was entered into a two-way MANOVA with E2 treatment and type of stimulus as between-subjects factors. Spearman's correlation tests were used to look for relations between behavior and the neural variables quantified (TH-IR cell number, fiber density, and colocalization of TH- and ZENK-IR). Plasma E2 levels were compared between treatment groups using a t-test.

4.10.

Detecting ZENK-IR in A6 cells

In order to evaluate our ability to detect ZENK protein in A6 cells, a separate group of females was given subcutaneous injections of NMDA (25 mg/kg; n = 3) or saline (n = 3). NMDA treatment induces FOS-IR in A6 neurons in mice (Saitoh et al., 1991). 60 min following the injections, birds were killed and their brains were processed as above for ZENK and TH doublelabel immunocytochemistry, except that instead of anti-TH, an anti-DBH antibody (Immunostar; dilution of 1:1600) was used in order to definitively label A6. The A6 region was inspected using the 20× and 40× objectives on the microscope in order to assess double-labeling.

Acknowledgments We are grateful to Ellen Cho, Susie Lackey, Henry Lange, and Mark Wilson for technical assistance and advice, and to Robert Liu and Darryl Neill for comments on an earlier version of the manuscript. We also thank Rashidat Ayantungi, Tulasi Ghimirey, and Marsha Howard for expert animal care. This work was supported by NSF IBN-0346984, HHMI 52003727, and the Center for Behavioral Neuroscience.

REFERENCES

Appeltants, D., Absil, P., Balthazart, J., Ball, G.F., 2000. Identification of the origin of catecholaminergic inputs to HVc in canaries by retrograde tract tracing combined with tyrosine hydroxylase. J. Chem. Neuroanat. 18, 117–133. Appeltants, D., Ball, G.F., Balthazart, J., 2001. The distribution of tyrosine hydroxylase in the canary brain: demonstration of a specific and sexually dimorphic catecholaminergic innervation of the telencephalic song control nuclei. Cell Tissue Res. 304, 237–259. Appeltants, D., Ball, G.F., Balthazart, J., 2002a. The origin of

101

catecholaminergic inputs to the song control nucleus RA in canaries. NeuroReport 13, 649–653. Appeltants, D., Del Negro, C., Balthazart, J., 2002b. Noradrenergic control of auditory information processing in female canaries. Behav. Brain Res. 113, 221–235. Appeltants, D., Gentner, T.Q., Hulse, S.H., Balthazart, J., Ball, G.F., 2005. The effect of auditory distractors on song discrimination in male canaries (Serinus canaria). Behav. Processes 69, 331–341. Aston-Jones, G., Cohen, J.D., 2005. An integrative theory of locus coeruleus–norepinephrine function: adaptive gain and optimal performance. Annu. Rev. Neurosci. 28, 403–450. Bailey, D.J., Rosebush, J.C., Wade, J., 2002. The hippocampus and caudomedial neostriatum show selective responsiveness to conspecific song in the female zebra finch. J. Neurobiol. 52, 43–51. Bailhache, T., Balthazart, J., 1993. The catecholaminergic system of the quail brain: immunocytochemical studies of dopamine beta-hydroxylase and tyrosine hydroxylase. J. Comp. Neurol. 329, 230–256. Balthazart, J., Ball, G., 1996. Identification of catecholaminergic cell groups in the brainstem of the canary, zebra finch, white-throated sparrow and budgerigar by tyrosine hydroxylase immunocytochemistry. Belg. J. Zool. 126, 65–78. Barclay, S.R., Harding, C.F., 1988. Androstenedione modulation of monoamine levels and turnover in hypothalamic and vocal control nuclei in the male zebra finch: steroid effects on brain monoamines. Brain Res. 459, 333–343. Barclay, S.R., Harding, C.F., 1990. Differential modulation of monoamine levels and turnover rates by estrogen and/or androgen in hypothalamic and vocal control nuclei of male zebra finches. Brain Res. 523, 251–262. Bao, S., Chan, V.T., Merzenich, M., 2001. Cortical remodeling induced by activity of ventral tegmental dopamine neurons. Nature 412, 79–83. Beaudet, A., Descarries, L., 1978. The monoamine innervation of rat cerebral cortex: synaptic and nonsynaptic axon terminals. Neuroscience 3, 851–860. Bharati, I.S., Goodson, J.L., 2006. Fos responses of dopamine neurons to sociosexual stimuli in male zebra finches. Neuroscience 143, 661–670. Cabot, J.B., Reiner, A., Bogan, N., 1982. Avian bulbospinal pathways: anterograde and retrograde studies of cells of origin, funicular trajectories and laminar terminations. Prog. Brain Res. 57, 79–108. Cardin, J.A., Schmidt, M.F., 2004. Noradrenergic inputs mediate state dependence of auditory responses in the avian song system. J. Neurosci. 24, 7745–7753. Castelino, C.B., Ball, G.F., 2005. A role for norepinephrine in the regulation of context-dependent ZENK expression in male zebra finches (Taeniopygia guttata). Eur. J. Neurosci. 21, 1962–1972. Castelino, C.B., Diekamp, B., Ball, G.F., 2007. Noradrenergic projections to the song control nucleus area X of the medial striatum in male zebra finches (Taeniopygia guttata). J. Comp. Neurol. 502, 544–562. Foote, S.L., Aston-Jones, G., Bloom, F.E., 1980. Impulse activity of locus coeruleus neurons in awake rats and monkeys is a function of sensory stimulation and arousal. Proc. Natl. Acad. Sci. U. S. A. 77, 3033–3037. Gahr, M., Guttinger, H.R., Kroodsma, D.E., 1993. Estrogen receptors in the avian brain: survey reveals general distribution and forebrain areas unique to songbirds. J. Comp. Neurol. 327, 112–122. Grant, S.J., Aston-Jones, G., Redmond Jr., D.E., 1988. Responses of primate locus coeruleus neurons to simple and complex sensory stimuli. Brain Res. Bull. 21, 401–410. Gentner, T.Q., Hulse, S.H., Duffy, D., Ball, G.F., 2001. Response biases in auditory forebrain regions of female songbirds

102

BR A IN RE S E A RCH 1 1 71 ( 20 0 7 ) 9 3 –1 03

following exposure to sexually relevant variation in male song. J. Neurobiol. 46, 48–58. Griffiths, R., Double, M.C., Orr, K., Dawson, R.J., 1998. A DNA test to sex most birds. Mol. Ecol. 7, 1071–1075. Helena, C.V., de Oliveira Poletini, M., Sanvitto, G.L., Hayashi, S., Franci, C.R., Anselmo-Franci, J.A., 2006. Changes in alpha-estradiol receptor and progesterone receptor expression in the locus coeruleus and preoptic area throughout the rat estrous cycle. J. Endocrinol. 188, 155–165. Hurley, L.M., Devilbiss, D.M., Waterhouse, B.D., 2004. A matter of focus: monoaminergic modulation of stimulus coding in mammalian sensory networks. Curr. Opin. Neurobiol. 14, 488–495. Ivanova, T., Beyer, C., 2003. Estrogen regulates tyrosine hydroxylase expression in the neonate mouse midbrain. J. Neurobiol. 54, 638–647. Kern, M.D., King, J.R., 1972. Testosterone-induced singing in female white-crowned sparrows. Bird Band. 45, 224–252. Kritzer, M.F., 1997. Selective colocalization of immunoreactivity for intracellular gonadal hormone receptors and tyrosine hydroxylase in the ventral tegmental area, substantia nigra, and retrorubral fields in the rat. J. Comp. Neurol. 379, 247–260. Kritzer, M.F., Kohama, S.G., 1998. Ovarian hormones influence the morphology, distribution, and density of tyrosine hydroxylase immunoreactive axons in the dorsolateral prefrontal cortex of adult rhesus monkeys. J. Comp. Neurol. 395, 1–17. Kritzer, M.F., Kohama, S.G., 1999. Ovarian hormones differentially influence immunoreactivity for dopamine β-hydroxylase, choline acetyltransferase, and serotonin in the dorsolateral prefrontal cortex of adult rhesus monkeys. J. Comp. Neurol. 409, 438–451. Kuppers, E., Ivanova, T., Karolczak, M., Beyer, C., 2000. Estrogen: a multifunctional messenger to nigrostriatal dopaminergic neurons. J. Neurocytol. 29, 375–385. Leitner, S., Voigt, C., Metzdorf, R., Catchpole, C.K., 2005. Immediate early gene (ZENK, Arc) expression in the auditory forebrain of female canaries varies in response to male song quality. J. Neurobiol. 64, 275–284. Maney, D.L., Ball, G.F., 2003. Fos-like immunoreactivity in catecholaminergic brain nuclei after territorial behavior in free-living song sparrows. J. Neurobiol. 56, 163–170. Maney, D.L., Bernard, D.J., Ball, G.F., 2001. Gonadal steroid receptor mRNA in catecholaminergic nuclei of the canary brainstem. Neurosci. Lett. 311, 189–192. Maney, D.L., MacDougall-Shackleton, E.A., MacDougallShackleton, S.A., Ball, G.F., Hahn, T.P., 2003. Immediate early gene response to hearing song correlates with receptive behavior and depends on dialect in a female songbird. J. Comp. Physiol., A Sens. Neural Behav. Physiol. 189, 667–674. Maney, D.L., Erwin, K.L., Goode, C.T., 2005. Neuroendocrine correlates of behavioral polymorphism in white-throated sparrows. Horm. Behav. 48, 196–206. Maney, D.L., Cho, E., Goode, C.T., 2006. Estrogen-dependent selectivity of genomic responses to birdsong. Eur. J. Neurosci. 23, 1523–1529. Mello, C.V., Clayton, D.F., 1994. Song-induced ZENK gene expression in auditory pathways of songbird brain and its relation to the song control system. J. Neurosci. 14, 6652–6666. Mello, C.V., Ribeiro, S., 1998. ZENK protein regulation by song in the brain of songbirds. J. Comp. Neurol. 393, 426–438. Mello, C.V., Vicario, D.S., Clayton, D.F., 1992. Song presentation induces gene expression in the songbird forebrain. Proc. Natl. Acad. Sci. 89, 6818–6822. Mello, C.V., Pinaud, R., Ribeiro, S., 1998. Noradrenergic system of the zebra finch brain: immunocytochemical study of dopamine beta-hydroxylase. J. Comp. Neurol. 400, 207–228. Mello, C.V., Velho, T.A., Pinaud, R., 2004. Song-induced gene

expression: a window on song auditory processing and perception. Ann. N. Y. Acad. Sci. 1016, 263–281. Moore, M., 1983. Effect of female displays on the endocrine physiology and behavior of male white crowned sparrows, Zonotrichia leucophrys. J. Zool. 137–148. Pau, K.Y.F., Hess, D.L., Kohama, S., Bao, J., Pau, C.Y., Spies, H.G., 2000. Oestrogen upregulates noradrenaline release in the mediobasal hypothalamus and tyrosine hydroxylase gene expression in the brainstem of ovariectomized rhesus macaques. J. Neuroendocrinol. 12, 899–909. Pazol, K., Kaplan, J.R., Abbott, D., Wilson, M.E., 2004. Practical measurement of total and bioavailable estradiol in female macaques. Clin. Chim. Acta 340, 117–126. Reader, T.A., Ferron, A., Descarries, L., Jasper, H.H., 1979. Modulatory role for biogenic amines in the cerebral cortex. Microiontophoretic studies. Brain Res. 160, 217–229. Reiner, A., Karle, E.J., Anderson, K.D., Medina, L., 1994. Catecholaminergic perikarya and fibers in the avian nervous system. In: Smeets, W.J.A., Reiner, A. (Eds.), Phylogeny and Development of Catecholamine Systems in the CNS of Vertebrates. Cambridge Univ. Press, Cambridge, UK, pp. 135–181. Reiner, A., Perkel, D.J., Bruce, L.L, Butler, A.B., Csillag, A., Kuenzel, W., et al., 2004. Revised nomenclature for avian telencephalon and some related brainstem nuclei. J. Comp. Neurol. 473, 377–414. Ribeiro, S., Cecchi, G.A., Magnasco, M.O., Mello, C.V., 1998. Toward a song code: evidence for a syllabic representation in the canary brain. Neuron 21, 359–371. Riters, L.V., Pawlisch, B.A., 2007. Evidence that norepinephrine influences responses to male courtship song and activity within song control regions and the ventromedial nucleus of the hypothalamus in female European starlings. Brain Res. 1149, 127–140. Riters, L.V, Eens, M., Pinxten, R., Ball, G.F., 2002. Seasonal changes in the densities of alpha(2) noradrenergic receptors are inversely related to changes in testosterone and the volumes of song control nuclei in male European starlings. J. Comp. Neurol. 444, 63–74. Saitoh, Y., Silverman, A.J., Gibson, M.J., 1991. Norepinephrine neurons in mouse locus coeruleus express c-fos protein after N-methyl-D,L-aspartic acid (NMDA) treatment: relation to LH release. Brain Res. 561, 11–19. Shu, S., Gong, J., Fan, L., 1988. The glucose oxidase–DAB–nickel method in peroxidase histochemistry of the nervous system. Neurosci. Lett. 85, 169–171. Serova, L., Rivkin, M., Nakashima, A., Sabban, E.L., 2002. Estradiol stimulates gene expression of norepinephrine biosynthetic enzymes in rat locus coeruleus. Neuroendocrinology 75, 193–200. Serova, L.I., Maharjan, S., Huang, A., Sun, D., Kaley, G., Sabban, E.L., 2004. Response of tyrosine hydroxylase and GTP cyclohydrase I gene expression to estrogen in brain catecholaminergic regions varies with mode of administration. Brain Res. 1015, 1–8. Shank, M.C., 1959. The natural termination of the refractory period in the slate-colored junco and in the white-throated sparrow. Auk 76, 44–54. Sockman, K.W., Gentner, T.Q., Ball, G.F., 2002. Recent experience modulates forebrain gene-expression in response to mate-choice cues in European starlings. Proc. R. Soc. Lond. B. 269, 2479–2485. Stokes, T.M., Leonard, C.M., Nottebohm, F., 1974. The telencephalon, diencephalon, and mesencephalon of the canary, Serinus canaria, in stereotaxic coordinates. J. Comp. Neurol. 156, 337–374. Stripling, R., Kruse, A.A., Clayton, D.F., 2001. Development of song responses in the zebra finch caudomedial neostriatum: role of genomic and electrophysiological activities. J. Neurobiol. 48, 163–180.

BR A I N R ES E A RC H 1 1 7 1 ( 2 00 7 ) 9 3 –1 03

Terleph, T.A., Mello, C.V., Vicario, D.S., 2006. Auditory topography and temporal response dynamics of canary caudal telencephalon. J. Neurobiol. 66, 281–292. Terpstra, N.J., Bolhuis, J.J., Riebel, K., van der Burg, J.M., den Boer-Visser, A.M., 2006. Localized brain activation specific

103

to auditory memory in a female songbird. J. Comp. Neurol. 494, 784–791. Wolfson, A., 1958. Regulation of refractory period in the photoperiodic responses of the white-throated sparrow. J. Exp. Zool. 139, 349–379.