Effects of social cues on GnRH-I, GnRH-II, and reproductive physiology in female house sparrows (Passer domesticus)

Transcription

1 vailable online at General and Comparative Endocrinology 156 (28) Effects of social cues on GnRH-I, GnRH-II, and reproductive physiology in female house sparrows (Passer domesticus) Tyler J. Stevenson a,, George E. Bentley b, Takayoshi Ubuka b, Lutgarde rckens c, Elizabeth Hampson a, Scott. MacDougall-Shackleton a a Department of Psychology and Graduate Program in Neuroscience, University of Western Ontario, US b Department of Integrative Biology and Helen Wills Neuroscience Institute, University of California, Berkeley, US c Department of nimal Physiology and Neurobiology, K.U. Leuven, Belgium Received 1 October 27; revised 21 December 27; accepted 8 January 28 vailable online 26 January 28 bstract In all vertebrates, at least two forms of gonadotropin-releasing hormone (GnRH) are present: GnRH-I and GnRH-II. GnRH-I directly influences the reproductive axis whereas the function of GnRH-II is less clear. The present experimental objectives were to determine the effect(s) of male social cues on the peripheral and neural responses of female house sparrows (Passer domesticus). We hypothesized that male breeding status would significantly influence the amount of immunoreactive GnRH-II in female house sparrow brains. In order to test this hypothesis, females were caged with a breeding male, a non-breeding male, or caged alone. The presence of breeding males did not significantly influence ovary development, luteinizing hormone, or estradiol levels, but male presence increased female body mass, and male presence and condition interacted to influence ovarian follicle size. Using immunocytochemistry, GnRH-I and GnRH-II immunoreactivity was measured in order to evaluate the neuroendocrine response to breeding status in males. When females were housed with breeding males, there were stable numbers of immunoreactive GnRH-I and -II cells but significantly lower amounts of immunoreactive GnRH-I fibre staining within the preoptic area compared to females housed with non-breeding males. Moreover, immunoreactive GnRH-II fibres in the preoptic area, ventromedial nucleus, and medial septum were significantly greater in females housed alone in chamber with non-breeding males. The data demonstrate that the GnRH system in songbirds is modulated by social context. These finding provide novel insight into the mechanisms involved with regulating avian reproductive physiology. Ó 28 Elsevier Inc. ll rights reserved. Keywords: GnRH-II; LHRH; Reproduction; ; Social cues; Immunocytochemistry 1. Introduction Corresponding author. Present address: Department of Psychological and Brain Sciences, The Johns Hopkins University, 34 N. Charles Street, Baltimore, MD 21218, US. address: tsteve13@jhu.edu (T.J. Stevenson). Social cues have a wide variety of effects on physiology. In many seasonally breeding animals photoperiod is the primary factor that controls seasonal changes in reproductive physiology, but social cues are critical for the fine tuning of reproduction. Social cues can influence peripheral reproductive physiology such as hormone levels and thus modify hormone-dependent behaviors (e.g., Hinde, 1965; Hinde and Steel, 1976). In addition, social cues can influence central physiology by modifying gonadotropin-releasing hormone (GnRH) neurosecretory cells through increased synthesis in the brain (e.g., White et al., 22; Burmeister and Wilczynski, 25). In this study, we explore the effects of visual and acoustic social cues on central and peripheral reproductive physiology of female house sparrows (Passer domesticus) /$ - see front matter Ó 28 Elsevier Inc. ll rights reserved. doi:1.116/j.ygcen

2 386 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) The effects of vocalizations on reproductive physiology have been long studied in birds. Studies in ring doves (Streptopelia risoria) primarily investigated the interaction between social context and the physiological response, focusing on the subsequent effect on breeding behavior (Lehrman, 1959, 1965). Lehrman demonstrated that male presence and nesting material significantly affected the courtship, copulatory, and incubating behavior of female ring doves (Lehrman et al., 1961a,b; Lehrman and Wortis, 196). Furthermore, Cheng et al., 1998 demonstrated that the females own behavior stimulated the hypothalamuspituitary-gonadal axis. In canaries, male vocalizations were found to accelerate and synchronize reproductive physiology and behaviors for rapid sexual development (Hinde and Steel, 1976). Specifically, female canaries were found to engage in nest building behaviors more readily when exposed to tape-recorded male canary song. These authors proposed that nest building and nest occupation by females are generally under the control of estrogen, and that male song and photostimulation act in a synergistic manner for reproductive maturation (Hinde and Steel, 1976). Perhaps because songbirds are such a popular study subject in behavioral ecological studies on sexual selection, research on the effects of male song on female hormones and behavior has continued (e.g., Kroodsma, 1976; Morton et al., 1985; Bentley et al., 2; MacDougall-Shackleton et al., 21; Cheng et al., 1998). However, the effects of other social cues, such as plumage, beak color, or other visual displays on female hormone levels have been less well studied. In addition, the effects of social cues on the GnRH system have been understudied in birds compared to other taxa such as frogs (Burmeister and Wilczynski, 25) or fish (White et al., 22), though recent data from ring doves provide an exception (Cheng et al., 1998). House sparrows are highly social seasonally breeding songbirds. This species exhibits seasonal changes in singing rates and beak color in association with changes in reproductive physiology (Hegner and Wingfield, 1986). In addition this species exhibits plasticity in both GnRH-I (Hahn and Ball, 1995) and GnRH-II (Stevenson and MacDougall-Shackleton, 25) in association with photoperiod-driven changes in reproductive condition. In this study, we investigated the effects of male presence and breeding on the reproductive physiology and neural GnRH-I and -II in female house sparrows using a 2 2 (male presence by breeding condition) experimental designs. 2. Methods 2.1. Subjects Wild house sparrows (N = 12 males, 24 females) were caught between July and October 24. Using mist nets, birds were captured from rural Southwestern Ontario. fter capture, birds were transported to the University of Western Ontario and group housed in an outdoor aviary. Birds were treated and handled in accordance with all appropriate animal care guidelines and permits Treatment groups On November 18, all birds were brought into the laboratory and randomly assigned (2 females, 1 male) into isolation chambers and kept on the natural photoperiod (9L:15D). Within each isolation chamber, one female was placed with the male in one cage and the other female housed in a second cage. On November 25 (Day ), all chambers were placed on a long day photoperiod (15L:9D) to induce breeding condition and remained on this photoperiod until the termination of the experiment. In order to manipulate social cues produced by males we performed castration and sham surgeries, and placed females either in a cage with a male or in a separate cage behind an opaque partition. For the castration surgeries, six males prior to photostimulation were anesthetized with isoflurane vapors and a small incision was made in the lower abdomen and the testes were removed with forceps. For the sham surgeries six males were anesthetized with isoflurane vapors and a small incision was made in the lower abdomen and the testes were touched with forceps but not removed. These two manipulations (castration, placement of a partition) created four treatment groups: (1) females housed with a breeding condition male; (2) females housed in a separate cage in the same chamber as the breeding male; (3) females housed with a non-breeding condition male; and (4) females housed in a separate cage in the same chamber as a non-breeding condition male. Females housed with a male (groups 1 and 3) were in the same cage as the male and could fully interact with him. Females housed behind the partition (groups 2 and 4), but in the same chamber, could vocally interact with the male but could not see him. Similarly, females within a chamber could hear, but not see each other ssessment of male reproductive condition To determine male breeding condition we measured beak color, song rate, and plasma testosterone concentrations. Black beaks indicate elevated levels of plasma androgens associated with the breeding season in this species (Pfeiffer et al., 1944; Hegner and Wingfield, 1986). Beak color was assessed on Day 1 and Day 12. Male house sparrows produce songs that consist of a single element called a cheep (Lowther and Cink, 1992). Song rates were collected beginning on Day and continued for 15 days. Songs were recorded from two randomly selected males every morning for about 1 h beginning when the lights were turned on. Recorded songs were later counted and singing rate was computed by dividing total song count by minutes recorded. Blood samples were collected on Day 1 and Day 12 to measure circulating testosterone concentrations. Blood samples were collected in heparinized microhematocrit tubes from a wing vein after puncture with a 26-ga needle. pproximately 3 4 ll of whole blood were collected from each male within 15 min or less of opening the chamber door. Blood samples were then centrifuged for 1 min, supernatant plasma was harvested with a Hamilton microsyringe, and stored at 2 C until radioimmunoassay. The serum was analyzed in duplicate (5 ll), without extraction, using a commercially available 125 I Coat--Count kit for total testosterone (Diagnostic Products Corporation, Los ngeles, C). The antiserum is highly specific for testosterone and shows negligible cross-reactivity with other steroids, including dihydrotestosterone (<5%). For greater sensitivity, a 1 ng standard was added to the curve. The sensitivity of the assay was less than 1 ng/dl and the intra-assay coefficient of variation averaged 7% ssessment of female reproductive condition We assessed female reproductive response by measuring body mass, largest ovarian follicle volume, ovary stage, hormones, and neural immunoreactivity for GnRH-I and -II. Body mass was measured to the nearest.1 g on Day 3, Day 4, Day 22, and Day 27. t the end of the experiment the birds were killed for brain analysis (see below) and the ovaries were dissected out and the largest follicle was measured to the nearest.1 mm using dial calipers. Follicle volume was calculated using the equation for spherical volume (4/3 pr 3 ), where r is half the diameter of the largest follicle. dditionally, ovary stage was assessed using the following scale: (1)

3 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) smooth, no visible follicular development; (2) slightly granular appearance; (3) small follicles apparent, but no follicular hierarchy; (4) obvious follicles with evident hierarchy; and (5) large yolky follicles. To measure luteinizing hormone (LH) and estradiol (E2) we collected blood samples from females on Day 3, Day 4, Day 22, and Day 27. Samples collected on Day 22 were assayed for E2, and the other samples were assayed for LH. Blood collection for females was otherwise identical to that for males (see above). Plasma was assayed for LH using a homologous chicken LH radioimmunoassay (Follett et al., 1972). ll samples were run in duplicate in a single assay. Intra-assay coefficient of variation was 4.2%, and the detection limit was.39 ng/ml. For the estradiol assay the serum was brought to 4 C and 15 ll of plasma was extracted in diethyl ether and re-constituted in 4 ll of phosphate-buffered saline. The assay was performed using a single Coat--Count kit for estradiol (Diagnostic Products Corporation, Los ngeles, C) with the curve extended to include a 1 pg standard. The assay was performed in standard duplicates, using a 2 ll aliquot of the extract. Sensitivity was 1 pg/ml and the intra-assay coefficient of variation averaged 1.1% Perfusion and brain sectioning t the end of the experiment, birds were killed and brains removed in order to measure GnRH-I and -II immunoreactivity. Birds were deeply anaesthetized with xylazine and ketamine 1:1. Next, birds were transcardially perfused with heparinized.1 M phosphate-buffered saline (PBS) ph 7.5, followed by 4% paraformaldehyde. Then the brains were dissected out and placed in 4% paraformaldehyde and left overnight at 4 o C. The following morning, the brains were transferred to a sucrose solution (3% sucrose in.1 M PBS) and left overnight at 4 o C. The brains were then frozen with powdered dry ice for 2 min, then left in a freezer ( 7 o C) until sectioning. Brains were sectioned coronally (4 lm) using a cryostat and sections were collected immediately from the region of the tractus septomesencephalicus (TSM) split to the termination of the third cranial nerve. Every section was placed in a well containing.1 M PBS until immunocytochemistry processing with alternate sections labeled for GnRH-I and GnRH-II Immunocytochemistry (ICC) The primary antibodies used in this study are highly specific for cgnrh- I and cgnrh-ii (e.g., Stevenson and MacDougall-Shackleton, 25; van Gils et al., 1993). The peptides were raised in rabbit and specificity determined in quail. For this study, we wanted to further demonstrate the specificity of the GnRH-I and GnRH-II antibodies. Therefore, we conducted a dot immunoblot assay. In brief, aliquots of GnRH-I, GnRH-II, or lamprey GnRH-III peptides were spotted onto a.45-lm polyvinylidene difluoride membrane (PolyScreen Ò PVDF Hybridization Transfer Membrane; PerkinElmer Life nd nalytical Sciences, Inc., Waltham, M, US). The membrane was air dried at room temperature and baked for 3 min at 1 C to fix the peptides on the membrane. The membrane was washed for 1 min in.1 M Tris buffer (ph 7.5) with.1% Tween 2 and.9% NaCl and incubated for 6 min in blocking solution containing 5% skim milk in.1% Tween 2 and.9% NaCl (blocking buffer). fter blockage, the membrane was exposed for 6 min to GnRH-I or GnRH-II antibodies (1:1 dilution in blocking buffer). fter the primary immunoreaction, the membrane was then exposed for 6 min to anti-rabbit antibody linked to alkaline phosphatase (Vector Laboratories, Inc., Burlingame, C; 1:1 dilution in blocking buffer). fter washing, immunoreactive spots were detected by a reaction with nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate stock solution (Roche Diagnostics, Mannheim, Germany) in alkaline phosphatase buffer (.5 mm MgCl 2, 1 mm diethanolamine; ph 9.5; 1:5 dilution) for 2 3 min. The immunocytochemistry protocol used in this experiment was adopted from the protocol performed by Stevenson and MacDougall- Shackleton (25). Immunocytochemistry was carried out on all brains concurrently. Once the brains were sectioned, the sections were washed in.1 M PBS twice, once in.5% H 2 O 2 for 15 min., then washed three times in.1 M PBS and left overnight in normal goat serum (1% solution in.3% PBS/T [Triton X]) at 4 C. Sections were incubated in primary antibody (1:5 for cgnrh-i and 1:2 cgnrh-ii in.3% PBS/T) for 24 h. The sections were then washed twice in.1% PBS/T and incubated overnight. In the morning the sections were washed once more with.1% PBS/T, then incubated in biotinylated secondary antibody (goat anti-rabbit IgG, 1:25) for 1 h, washed three times in.1% PBS/T, incubated in avidin biotin horseradish-peroxidase complex (Vectastain BC, Elite Kit 1:2) for 1 h and then washed again three times in.1% PBS/T. The GnRH immunoreactive sites were visualized by incubating the sections with diaminobenzidine (Sigma Fast DB) for 9 s. Finally, sections were washed three times with.1 M PBS and were mounted onto gelatin coated microscope slides. Sections were then serially dehydrated in ethanol and then placed in citrosol (Fisher) for 5 min. The slides were coverslipped using Permount (Fisher) nalysis Immunoreactive cell bodies and fibres were measured using a brightfield light microscope. Images were captured using a Spot Insight digital camera mounted on a Zeiss xiophot microscope. The total numbers of ir-gnrh-i and -II cell bodies were counted manually. The density of irfibres for GnRH-I and -II in the preoptic area, medial septum and the ventromedial nucleus were measured using SigmaScan Pro 5. (SPSS Science). Immunoreactive GnRH fibre area was determined by converting captured images to grey scale and altering the illumination threshold to highlight only immunoreactive areas. The total number of highlighted pixels in each image was measured and averaged across 1 consecutive brain sections per bird. Fibre area was quantified separately for the PO, VMN, and medial septum. 3. Results 3.1. ntibody specificity for GnRH-I, GnRH-II antibodies The dot immunoassay demonstrated that the GnRH-I antibody only bound to GnRH-I peptide (Fig. 1) and the GnRH-II antibody only bound to GnRH-II peptide cgnrh-i antibody mole I II III 1 x x x x x x 1-15 B cgnrh-ii antibody mole I II III 5 x x x x x x 1-15 Fig. 1. Dot immunoassay demonstrating specificity of antibodies. () to mol of cgnrh-i (I), cgnrh-ii (II), or lamprey GnRH-III (III) peptides were spotted onto a polyvinylidene difluoride membrane. The membrane was incubated in cgnrh-i antibody (1:1 dilution) and the immunoreaction was visualized using alkaline phosphatase-conjugated secondary antibody. (B) to moles of cgnrh-i (I), cgnrh-ii (II), or lamprey GnRH-III (III) peptides were spotted onto a polyvinylidene difluoride membrane. The membrane was incubated in cgnrh-ii antibody (1:1 dilution) and the immunoreaction was visualized using alkaline phosphatase-conjugated secondary antibody.

4 388 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) (Fig. 1B). Furthermore, the GnRH-I antibody only labeled neurons in the preoptic area (Fig. 2) with no GnRH-II antibody labeled neurons in the PO (Fig. 2B) only in the mesencephalon central grey (Fig. 2C) Male breeding status Prior to photostimulation all males had pale yellow beaks indicating low T (Keck 1934). Following photostimulation two of the six males that had undergone castration surgery developed jet black beaks. Presumably, the castrations were incomplete or non-gonadal androgens were elevated. In addition, two of the six males that had undergone sham surgery did not develop black beaks. The reason for this is not clear, but perhaps they had not yet broken photorefractoriness (see Hahn and Ball, 1995). House sparrows exhibit large individual variation in the timing of photorefractoriness and a failure to respond to photoperiod manipulations by some individuals has been reported previously (Hahn and Ball, 1995). Because of this, rather than use castrate and sham to distinguish our treatment groups, we objectively classified males as being in breeding or non-breeding condition based on beak color. Six males with black beaks were categorized as being in breeding condition and six males with pale yellow beaks were categorized as being in non-breeding condition. Based on these categories, we evaluated the differences between testosterone and singing rates of breeding and non-breeding males. We conducted separate statistical analyses with the exclusion of the four male sparrows. Exclusion of these males did not qualitatively affect the results: all trends were in the same direction though some effects (GnRH-II fibre density in VMN, GnRH-I immunoreactivity) dropped slightly above statistical significance. Because excluding these birds lowers the sample size and also reduces statistical power, we decide to include the four males and consequently, eight female sparrows for analyses reported below. Despite clear differences in beak color, we did not detect differences between breeding and non-breeding males in testosterone or singing rates. Testosterone showed large variation between breeding and non-breeding males during the course of the experiment (Fig. 3). However, a two-way NOV found no significant difference between breeding and non-breeding males (F(1, 1) = 1.39, P =.26) and there was no significant affect of photoperiod (F(1, 1) = 1.1, P =.33). There was also no significant difference in singing rate between breeding and non-breeding males (t(1) = 2.15, P =.36, Fig. 3). lthough there were differences in beak color, the large within group variance in testosterone and singing rates (Fig. 3) likely limited ability to detect between group differences Female reproductive physiology Fig. 2. Specificity of immunoreactivity for GnRH-I and GnRH-II neurons. () GnRH-I antibody labeled neurons in the PO; (B) Lack of GnRH-II antibody labeled neurons in PO; (C) GnRH-II antibody labeled neurons in the mesencephalon central grey Body mass Females caged with a male gained body mass, whereas females caged alone tended to lose body mass for the duration of the experiment (Fig. 4). We conducted a three-way repeated measure NOV to determine the effect of male presence and breeding condition on female body mass. Male presence had a significant effect on female body mass (F(1, 2) = 7.31, P <.5) and body mass was significantly different across treatment days (F(3, 6) = 5.54, P <.5). Furthermore, there was a significant interaction between male presence and treatment period (F(3,6) = 5.15,

5 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) Testosterone (ng/dl) Males Male Body Mass (g) With Male With Male Not With Male Not With Male 22.5 B Song Rate (songs/min) Day -1 Day 12 Treatment Day Males Males B Follicle Volume (mm 3 ) Day -3 Day 9 Day 2 Day 27 Treatment Day With Male Not With Male Fig. 3. Testosterone and singing rates of males with black beaks (breeding condition) or with yellow beaks (non-breeding condition). Error bars indicate SEM. P <.5). Post hoc tests (Tukey) indicated that after Day 4, females housed with males were significantly heavier compared to females housed alone. There was no significant main effect of male breeding condition on female body mass (F(1, 2) =.41, P =.52) Ovaries There was large variation in the size of the largest follicle at the termination of the experiment (Fig. 4). two-way NOV showed a significant interaction for treatment group and male presence (F(1, 2) = 7.21, P <.5). However, there were no significant main effects of male presence (F(1, 2) =.2, P =.67) or breeding status (F(1, 2) =.75, P =.39). There were no significant differences in ovary stage between females that were housed with a male compared to females housed alone (F(1,2) =.22, P =.64), no significant effect of male breeding condition (F(1, 2) =.16, P =.69) nor an interaction (F(3, 2) = 2.83, P =.11) Luteinizing hormone There was a significant effect of photoperiod on LH (Fig. 5). three-way repeated-measures NOV indicated C Ovary Stage Fig. 4. Effects of male social cues on female reproductive physiology. () Mean (± SEM) body mass of females housed with a male in breeding or non-breeding condition or behind a partition in the same chamber as a male in breeding or non-breeding condition. (B) Mean (± SEM) volume of the largest ovarian follicle at the end of the experiment. (C) Mean (± SEM) ovary stage at the end of the experiment. sterisks indicate P <.5. a significant main effect of blood sampling day (F(2, 4) = 17.42, P <.1). Post hoc tests indicated that LH levels were significantly different between Day 3 and Day 4 (P <.1); Day 4 and Day 22 (P <.5); and Day 3 and Day 22 (P <.1). However, neither male

6 39 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) B Luteinizing Hormone (pg/ml) Estradiol Concentrations (pg/ml) With Male With Male Not With Male Not With Male Day -3 Day 4 Day 22 Treatment Day breeding condition (F(1, 2) = 1.98, P =.17); male presence (F(1,2) =.91; P =.35) nor an interaction (F(1, 2) =.76, P =.39) were significant. Furthermore, there were no significant interactions of LH and male presence (F(2, 4) =.19, P =.83); breeding condition (F(2,4) = 1.16, P =.32); or a three-way interaction (F(2,4) = 1.4, P =.36) Estradiol t the termination of the experiment (Day 22), females had plasma estradiol concentrations indicative of a breeding state (Fig. 5). two-way NOV revealed no significant main effect of male presence (F(1, 2) = 1.46, P =.24) or breeding condition (F(1,2) = 1.29, P =.27), nor a significant interaction (F(1, 2) = 1.39, P =.25) Female neural GnRH-I and -II With Male Not With Male Fig. 5. Effects of male social cues on female hormone levels. () Mean (± SEM) plasma LH of females housed with a male in breeding or nonbreeding condition or behind a partition in the same chamber as a male in breeding or non-breeding condition. (B) Mean (± SEM) plasma estradiol. sterisks indicate P <.5. The neural distribution of immunoreactive GnRH-I and - II observed was similar to that reported in previous studies (van Gils et al., 1993; Millam et al., 1993; Stevenson and MacDougall-Shackleton, 25; Stevenson et al., 27). GnRH-I cells were primarily located in the PO, commencing caudal to the TSM split and continued toward the anterior commissure. Immunoreactive GnRH-I fibres were identified in many hypothalamic nuclei but primarily in the PO and VMN. dditional immunoreactive fibre staining was observed in the septum. GnRH-II cell bodies were distributed along the ventricle in close proximity to the third nerve in the mesencephalic central grey in the oculomotor complex. These cell bodies appeared to project fibres primarily into the PO with less dense projections into the VMN, lateral hypothalamus, median eminence, habenula, septal regions, and hippocampus. Immunoreactive GnRH-I and - II fibre staining exhibited considerable variation between female treatment groups. Fig. 6 represents the general immunoreactive fibre staining patterns observed in females exposed to the different treatment groups GnRH-I immunoreactivity Male social cues did not affect the number of cells immunoreactive for GnRH-I (Fig. 7). two-way NOV revealed no main effect of male presence (F(1, 2) =.85, P =.36). Furthermore, there was no significant main effect of male breeding status (F(1,2) =.69, P =.41) or a significant interaction (F(1, 2) = 1.55, P =.99). Immunoreactive GnRH-I fibre area within the PO exhibited large variation (Fig. 7). Unlike the cell numbers, females housed with a breeding male had lower immunoreactive GnRH-I fibre staining compared to females housed with a non-breeding male (F(1, 2) = 7.98, P <.5). However, neither male presence (F(1, 2) =.38, P =.54) nor the interaction (F(1, 2) =.17, P =.69) were significant GnRH-II immunoreactivity Similar to immunoreactive GnRH-I cell bodies, male social cues did not appear to affect the number of cells immunoreactive for GnRH-II (Fig. 8). There were no significant main effects of male breeding condition (F(1, 2) =.69, P =.41) or male presence (F(1, 2) =.85, P =.36) nor a significant interaction (F(1, 2) =.42, P =.53). Furthermore, there were no significant affects of treatment groups on GnRH-II cell size (F(1,2) = 2.45, P =.9). In contrast to the number of immunoreactive GnRH-II cells, there was significant variation among treatment groups in immunoreactive GnRH-II fibre area. Females housed with a breeding male had significantly fewer immunoreactive GnRH-II fibres in the PO (F(1, 2) = 6.32, P <.5), VMN (F(1, 2) = 4.45, P <.5), and medial septum (F(1, 2) = 19.63, P <.1). Furthermore, there was a significant main effect of male presence on the fibre area of GnRH-II in the medial septum (F(1, 2) = 7.66, P <.5) but not in the PO (F(1, 2) = 1.67, P =.21), or VMN (F(1, 2) = 3.16, P =.9). There was a significant interaction found for the medial septum (F(1,2) = 9.46, P <.1) but, not for the PO (F(1, 2) = 3.71, P =.6) or the VMN (F(1, 2) = 2.2, P =.17).

7 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) Fig. 6. Photomicrographs of GnRH-I and GnRH-II immunoreactivity in house sparrow females housed with a male in breeding or non-breeding condition or behind a partition in the same chamber as a male in breeding or non-breeding condition. 4. Discussion In this study, male and female house sparrows were placed on long day photoperiods to stimulate reproductive physiology and behavior. In order to test the hypothesis that male social cues would augment the development of reproductive physiology in females, we investigated the effect of male presence and breeding condition on ovarian development, luteinizing hormone, estradiol, and neural GnRH-I and -II. Through these manipulations, we demonstrated that (i) body mass was affected by male presence, (ii) ovarian follicle development was affected by a combination of male presence and breeding condition, and (iii) immunoreactive GnRH-I and -II fibre area was affected by the breeding condition of potential mates Male breeding condition and female reproductive physiology Body mass The onset of breeding and subsequent egg-laying is stimulated by metabolic changes (Lehikoinen, 1987) that lead to an increase in body mass. Increases in photoperiod have been demonstrated to increase body mass in Japanese quail (Boon et al., 2) and white-crowned sparrows (Morton et al., 24). Recently, Sockman et al. (24) demonstrated that females with prior experience with photostimulation increased body mass more rapidly than naïve females. In the present study, we found that females housed with males had significant increases in body mass throughout the experiment while females housed alone decreased body mass. These results suggest that an increase in photoperiod and mate social interactions resulted in body mass gain. n alternative hypothesis is that photostimulation increases body mass provided females are housed with another bird and not necessarily a potential mate. Regardless, an increase in female body mass may be a function of prebreeding hyperphagia Ovarian development Gonadal development is primarily dependent on an increase in vernal photoperiod; other various environmental stimuli modify this response. For example, in song sparrows that live in mountainous regions, cold temperatures may delay gonadal recrudescence by up to two months compared to coastal birds (Perfito et al., 24, 25). lso, the presence of nestboxes increased testis width in male European starlings (Sturnus vulgaris; Gwinner et al., 1988). Song cues produced by males stimulate the reproductive axis of female white-crowned sparrows (Morton

8 392 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) B GnRH-I PO fibre area (pixels) Number GnRH-I cell bodies With Male Not With Male 3x1 3 25x1 3 2x1 3 15x1 3 1x1 3 5x1 3 Fig. 7. Effects of male social cues on female neural GnRH-I immunoreactivity. () Mean (± SEM) GnRH-I cell counts within the PO. (B) Mean (± SEM) pixel area of immunoreactive GnRH-I fibres within the PO. sterisks indicate P <.5. et al., 1985; MacDougall-Shackleton et al., 21) and canaries (Serinus canarius; Bentley et al., 2); females in these studies exposed to conspecific song were found to possess significantly larger follicle volumes. Here we measured the largest follicle volume and ovarian developmental stage in response to the presence of males and the males breeding condition. lthough we found no effects of male presence or breeding condition on overall ovary stage, there was a significant interaction between male presence and breeding condition on the size of the largest ovarian follicle. These results are consistent with the hypothesis that photoperiod controlled overall ovarian development, but that preparation for oviposition (development of largest follicle) was affected by a variety of stimulatory and inhibitory factors. For example, being caged with a nonbreeding condition male (pale beak) may have inhibited follicular growth whereas being housed with a breeding condition male (black beak) may have accelerated follicular growth. Resolving this interaction clearly requires further studies Hormone levels Other reproductive physiological measures included LH and estradiol levels. In this study, circulating plasma LH increased after exposure to long days, indicating that females were photostimulated. We did not find a significant effect of male presence or breeding condition on female hormone concentrations. Experiments investigating the effect of male song on female reproductive physiology have identified effects on follicular development but not gonadotropin concentrations. Neither natal nor foreign song playbacks affect circulating LH levels in female white-crowned sparrows (MacDougall-Shackleton et al., 21). However, after 43 days of conspecific song playback, female canaries had higher levels of circulating LH compared to those who heard heterospecific or no song (Bentley et al., 2; Cheng et al., 1998). In the present study we did not detect any significant effects of male social cues on female hormone levels. Resolving whether these null results reflect a true lack of effect or poor statistical power requires further experiments. Detecting effects of social cues in females has generally proved more difficult than detecting similar effects in males (e.g., Moore, 1983; Pinxten et al., 23) Male breeding status and female neural GnRH Reproductive physiology is under the direct control of hypothalamic GnRH. The primary hypophysiotropic form, GnRH-I, is released at the median eminence to control release of gonadotropins from the pituitary (Sharp et al., 199). second form, GnRH-II, initially discovered in birds, has widespread immunoreactive fibre distribution in the brain with highly localized cell bodies (van Gils et al., 1993; Millam et al., 1993). We present additional evidence utilizing dot-blot to confirmed that the GnRH-I antiserum did not cross-react with GnRH-II, the GnRH-II antiserum did not cross-react with GnRH-I and neither antiserum cross-reacted with GnRH-III. The role of GnRH-II in modulating gonadotropin release is less clear (Sharp et al., 199) though a functional role for mediating sexual behavior has been implicated in both birds and mammals (Maney et al. 1997; Kaufmann, 24; Bentley et al., 26). In house sparrows, both GnRH-I and -II exhibit seasonal changes (Hahn and Ball, 1995; Stevenson and MacDougall-Shackleton, 25) in association with other changes in reproductive physiology. We thus tested whether GnRH-I and -II immunoreactivity varied in response to changes in male presence and breeding condition. Male presence and breeding condition affected the amount of neural GnRH-I and -II in females, as measured by immunocytochemistry. We found that male breeding condition significantly decreased the amount of immunoreactive GnRH-I fibre area in the PO and GnRH-II fibre area in the PO, VMN, and medial septum. There was no significant effect of male presence or breeding condition on the numbers of immunoreactive GnRH-I or -II cell bodies. We interpret the stable numbers of immunoreactive GnRH cells but decreased area of fibre immunoreactivity as resulting from increased release of peptide with relatively stable levels of synthesis. lternatively, the observed decrease in fibre immunoreactivity could also mean a

9 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) With Male Not With Male B 1 Number GnRH-II cell bodies Mean GnRH-II Cell Size (μm) Non- Male C GnRH-II PO fibre area (pixels) D GnRH-II VMN fibre area (pixels) E GnRH-II Septum fibre area (pixels) Fig. 8. Effects of male social cues on female neural GnRH-II immunoreactivity. () Mean (± SEM) GnRH-II cell counts within the oculomotor region of the mesencephalon. (B) Mean (± SEM) GnRH-II cell size in the mesencephalic central grey in the oculomotor complex. Mean (± SEM) pixel area of immunoreactive GnRH-II fibres within the (C) PO, (D) VMN, and (E) medial septum. sterisks indicate P <.5. decreased in synthesis and release. This hypothesis is further supported by the finding that cell area was not significantly affected by the treatment groups (data not presented). Thus females in a chamber with a breeding condition male may have been releasing more GnRH-I. This would suggest that male vocal cues are primarily responsible for this effect, since both females housed in the same cage as the breeding condition males and those behind a partition exhibited decreased GnRH-I fibre immunoreactivity (Fig. 7). These findings are intriguing and future research should investigate the mechanism involved with regulating the contextual modulation of GnRH-I in the PO. dditionally, the GnRH-II fibre immunoreactivity was also lower in all females except those housed behind a partition in a chamber with a non-breeding condition male (Fig. 8). If a decrease in immunoreactivity indicates increased release of GnRH, then the latter result would suggest that increased GnRH-II release may result from a greater variety of social cues. This is supported from data demonstrating a reduction of GnRH-II after social isolation (Perfito et al. 26). Seasonally breeding temperate zone birds primarily rely on the vernal increase in photoperiod to successfully optimize the breeding season (Dawson et al., 21). In addition to the increase in photoperiod, various supplementary cues are integrated resulting in the fine tuning to time reproductive success. In the present study, we investigated the effect of male breeding condition and presence on reproductive physiology in female house sparrows. We propose that social interactions may in part be regulated by the neural

10 394 T.J. Stevenson et al. / General and Comparative Endocrinology 156 (28) activity of GnRH-II functioning as a neuromodulator or neurotransmitter for reproductive behaviors. References Bentley, G.E., Wingfield, J.C., Morton, M.L., Ball, G.F., 2. Stimulatory effects on the reproductive axis in female songbirds by conspecific and heterospecific male song. Horm. Behav. 37, Bentley, G.E., Jensen, J.P., Kaur, G.J., Wacker, D.W., Tsutsui, K., Wingfield, J.C., 26. Rapid inhibition of female sexual behavior by gonadotropin-inhibitory hormone (GnIH). Horm. Behav. 49, Boon, B., Visser, G.H., Daan, S., 2. Effect of photoperiod on body mass gain, and daily energy intake and energy expenditure in Japanese quail (Coturnix c Japonica). Physiol. Behav. 7, Burmeister, S.S., Wilczynski, W., 25. Social signals regulate gonadotropin-releasing hormone neurons in the green treefrog. Brain Behav. Evol. 65, Cheng, M.F., Peng, J.P., Johnson, P., Hypothalamic neurons preferentially respond to female nest coo stimulation: demonstration of direct acoustic stimulation of luteinizing hormone release. J. Neurosci. 18, Dawson,., King, V.M., Bentley, G.E., Ball, G.F., 21. Photoperiodic control of seasonality in birds. J. Biol. Rhythms 16, Follett, B.K., Scanes, C.G., Cunningham, F.J., radioimmunoassay for avian luteinizing hormone. J. Endocrinol. 52, Gwinner, H., Gwinner, E., Dittami, J., Effect of nestboxes on LH, testosterone, testicular size and the reproductive behavior of male European starlings in spring. Behavior 13, Hahn, T.P., Ball, G.F., Changes in brain GnRH associated with photorefractoriness in house sparrows (Passer domesticus). Gen. Comp. Endocrinol. 99, Hegner, R.E., Wingfield, J.C., Gonadal development during autumn and winter in house sparrows. Condor 88, Hinde, R.., Interaction of internal and external factors in integration of canary reproduction. In: Beach, F.. (Ed.), Sex and Behavior. Wiley, New York, pp Hinde, R.., Steel, E., The effect of male song on an estrogendependent behavior pattern in the female canary (Serinus canarius). Horm. Behav. 7, Kaufmann,.S., 24. Emerging functions of gonadotropin-releasing hormone II in mammalian physiology and behavior. J. Neuroendocrinol. 16, Keck, W.N., Control of the sex characters in the English sparrow, Passer domesticus (Linnaeus). Proc. Soc. Exp. Biol. Med. 3, Kroodsma, D.E., Reproductive development in the female songbird: differential stimulation by quality of male song. Science 192, Lehikoinen, E., Seasonality of the daily weight cycle in wintering passerines and its consequences. Ornis Scand. 18, Lehrman, D.S., Hormonal responses to external stimuli in birds. Ibis 11, Lehrman, D.S., Wortis, R.P., 196. Previous breeding experience and hormone-induced incubation behavior in the ring dove. Science 132, Lehrman, D.S., Brody, P.N., Wortis, R.P., 1961a. The presence of the mate and of nesting material as stimuli for the development of incubation behavior and for gonadotropin secretion in the ring dove (Streptopelia risoria). Endocrinology 68, Lehrman, D.S., Brody, P., Wortis, R.P., 1961b. Gonadotropin secretion in response to external stimuli of varying duration in the ring dove (Streptopelia risoria). Proc Soc. Exp. Biol. Med. 16, Lehrman, D.S., Interaction between internal and external environments in the regulation of the reproductive cycle of the ring dove. In: Beach, F.. (Ed.), Sex and Behavior. Wiley, New York, pp Lowther, P.E., Cink, C.L., House sparrow (Passer domesticus). In: Poole,., Stetenheim, P., Gill, F. (Eds.), The Birds of North merica. cademy of National Sciences, Philadelphia (No. 12). MacDougall-Shackleton, S.., MacDougall-Shackleton, E.., Hahn, T.P., 21. Physiological and behavioral responses of female mountain white-crowned sparrows to natal- and foreign-dialect songs. Can. J. Zool. 79, Maney, D.L., Richardson, R.D., Wingfield, J.C., Central administration of chicken gonadotropin-releasing hormone-ii enhances courtship behaviour in a female sparrow. Horm. Behav. 32, Millam, J.R., Faris, P.L., Youngren, O.M., Halawani, M.E., Hartman, B.K., Immunohistochemical localization of chicken gonadotropin-releasing hormones-i and II (GnRH-I and II) in turkey hen brain. J. Comp. Neurol. 333, Moore, M.C., Effect of female sexual displays on the endocrine physiology and behavior of male white crowned sparrows, Zonotrichia leucophrys. J. Zool. 199, Morton, M.L., Pereyra, M.E., Baptista, L.F., Photoperiodically induced ovarian growth in the white-crowned sparrow (Zonotrichia leucophrys gambelii) and its augmentation by song. Comp. Biochem. Physiol. 8, Morton, M.L., Pereyra, M.E., Crandall, J.D., MacDougall-Shackleton, E., Hahn, T.P., 24. Reproductive effort and return rates in the mountain white-crowned sparrow. Condor 16, Perfito, N., Tramontin,.D., Meddle, S.L., Sharp, P.J., fik, D., Gee, J., Ishii, S., Kikuchi, M., Wingfield, J.C., 24. Reproductive development according to elevation in seasonally breeding male songbird. Oecologica 14, Perfito, N., Meddle, S.L., Tramontin,.D., Sharp, P.J., Wingfield, J.C., 25. Seasonal gonadal recrudescence in song sparrows: response to temperature cues. Gen. Comp. Endocrinol. 143, Perfito, N., Bentley, G.E., Hau, M., 26. Tonic activation of brain GnRH immunoreactivity despite reduction of peripheral reproductive parameters in opportunistically breeding zebra finches. Brain Behav. Evol. 67, Pfeiffer, C.., Hooker, C.W., Kirschbaum,., Deposition of pigment in the sparrow s bill in response to direct applications as a specific and quantitative test for androgen. Endocrinology 34, Pinxten, R., Ridder, E., Eens, M., 23. Female presence effects male behavior and testosterone levels in the European starling (Sturnus vulgaris). Horm. Behav. 44, Sharp, P.J., Talbot, R.T., Main, G.M., Dunn, I.C., Fraser, H.M., Huskisson, N.S., 199. Physiological roles of chicken LHRH-I and -II in the control of gonadotropin release in the domestic chicken. J. Endocrinol. 124, Sockman, K.W., Williams, T.D., Dawson,., Ball, G.F., 24. Prior experience with photostimulation enhances photo-induced reproductive development in female European starlings: a possible basis for the age-related increase in avian reproductive performance. Biol. Reprod. 71, Stevenson, T.J., MacDougall-Shackleton, S.., 25. Season and age related variation in neural GnRH-I and GnRH-II in house sparrows (Passer domesticus). Gen. Comp. Endocrinol. 143, Stevenson, T.J., rckens, L., MacDougall-Shackleton, S.., 27. Distribution of gonadotropin releasing-hormone-ii in the house sparrow brain. Gen. Comp. Endocrinol. 15, van Gils, J., bsil, P., Grauwels, L., Moons, L., Vandesande, F., Balthazart, J., Distribution of luteinizing hormone-releasing hormone I and II (LHRH-I and -II) in the quail and chicken brain as demonstrated with antibodies directed against synthetic peptides. J. Comp. Neurol. 334, White, S.., Nguyen, T., Fernald, R.D., 22. Social regulation of gonadotropin-releasing hormone. J. Exp. Biol. 25,