The role of androgen receptors in regulating territorial aggression in male song sparrows

Hormones and Behavior 57 (2010) 86–95

Contents lists available at ScienceDirect

Hormones and Behavior j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y h b e h

The role of androgen receptors in regulating territorial aggression in male song sparrows Todd S. Sperry ⁎, Douglas W. Wacker 1, John C. Wingfield 2 Department of Biology, University of Washington, Seattle, WA 98195, USA

a r t i c l e

i n f o

Article history: Received 19 March 2009 Revised 21 September 2009 Accepted 24 September 2009 Available online 30 September 2009 Keywords: Androgen receptor Flutamide Antiandrogen Aggression Territorial behavior Testosterone Bird Song sparrow

a b s t r a c t This paper examines the role that androgen receptors (ARs) play in modulating aggressive behavior in male song sparrows, Melospiza melodia morphna. Song sparrows are seasonally breeding, territorial birds that maintain year-round territories with male–female pair bonds formed during the spring breeding season. Plasma testosterone levels peak as territories are established and mates acquired. In late summer, testosterone levels fall and remain basal during the non-breeding season. We examined the role of ARs in regulating territorial aggression in captive song sparrows under short- and long-day conditions as well as just prior to, and at the start of the breading season in freely living birds using the nonsteroidal antiandrogen flutamide to block AR function. Birds were implanted with either empty or drug filled silastic implants for 18 to 42 days and then challenged with a novel male decoy to assess the individual birds level of male–male aggression. Freely living birds remained on their home territory and underwent a simulated territorial intrusion, whereas laboratory-held birds were assessed using a laboratory simulated territorial intrusion and remained in their home cage. Experimental treatment of male song sparrows decreased aggressive behavior during the pre-breeding life history substage (March–April) in freely living birds as well as in laboratoryheld birds under long-day (16L:8D) conditions. During the early breeding substage (April–May) there was no measurable effect of flutamide treatment on aggressive behavior, nor was there a difference in behavior in the (8L:16D) laboratory birds. This demonstrates that ARs are an important component of the neuroendocrine control of aggressive behavior. Given that flutamide only affected aggression during the pre-breeding substage and in LD birds, the results suggest that AR dependent control of aggressive behavior changes as song sparrow life history states change. © 2009 Elsevier Inc. All rights reserved.

Introduction There is a large body of evidence demonstrating that steroids, and in particular testosterone (T), are involved with the regulation of territorial aggression (Nelson, 1995; Wingfield et al., 1999). Within a reproductive context, aggression is activated by T (Simon et al., 1996; Wingfield et al., 1999) with circulating levels correlating with aggression only during socially unstable periods (i.e. the Challenge Hypothesis) (Wingfield et al., 1990). This relationship between T and aggression is present throughout the vertebrate classes (Nelson, 1995), but the mechanisms through which T exerts its effect are not fully understood. T can facilitate aggressive behavior primarily through either an androgenic or estrogenic pathway. The androgen receptor (AR) transcription factor is thought to be the principal target

⁎ Corresponding author. E-mail address: [email protected] (T.S. Sperry). 1 Current address: Centre for Integrative Physiology, College of Medicine and Veterinary Medicine, University of Edinburgh, Edinburgh EH8 9XD, UK. 2 Current address: Department of Neurobiology, Physiology and Behavior, The University of California at Davis, One Shields Avenue, Davis, CA 95616-8519, USA. 0018-506X/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.yhbeh.2009.09.015

for the androgenic pathway with T acting either directly or after being converted by the steroidogenic enzyme 5α-reductase to the nonaromatizable androgen 5α-dihydrotestosterone (DHT). T can also be converted to the primary estrogenic steroid 17β-estradiol (E2) by the enzyme aromatase thus influencing aggression via either the α or β estrogen receptors (ER). There is also increasing evidence that there are non-genomic androgen and estrogen receptors in addition to the classical nuclear transcription factors (see Thomas et al., 2006) providing several other possible targets through which T can act. The activation of the ER by E2 plays a central role in the activation of male aggressive behavior. In male mice, ER-α is the predominant ER target involved in the activation of aggressive behavior (Ogawa et al., 1998), whereas ER-β appears to be involved in inhibiting aggressive behavior during puberty and in young adult male mice (Nomura et al., 2002). In fact, based upon knockout studies, ER-α appears to be necessary for T induction of aggressive behavior (Ogawa et al., 1998). Although in birds there is no direct evidence for a role for either ER-α or -β, there is considerable evidence that E2 is necessary for males to exhibit territorial aggression (Soma et al., 2000a). This has been shown within reproductive (Schlinger and Callard, 1989; Soma et al., 2000a), as well as non-reproductive

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

contexts (Soma et al., 2000b), but results do vary across species (Canoine and Gwinner, 2002). There is strong evidence that in addition to the ER-mediated regulation of male aggressive behavior an androgen-mediated component is necessary for an animal to express a full suite of aggressive behaviors (reviewed in Simon, 2002). While there is evidence for a role for the primary androgens T and DHT in the regulation of aggressive behavior (e.g. Archawaranon and Wiley, 1988; Finney and Erpino, 1976; Ogawa et al., 1996), there is less known about AR mediated control. In fact, these studies suggest that the combination of estrogenic and androgenic steroids synergistically modulates aggressive behavior. However, the mechanisms through which androgens are acting are not well understood. The recent development of a complete AR knockout mouse has demonstrated that the absence of the AR protein results in reduced male territorial aggression (Sato et al., 2004). Thus, confirming that the AR is necessary for complete expression of aggressive behavior. However, this study also found that male aggression may be partially regulated through androgens acting via an AR-independent pathway distinct from the ER receptors (Sato et al., 2004). In non-mammalian vertebrates there is evidence of the importance of the AR mediated pathways using antiandrogens to prevent activation of the AR. Cyproterone acetate, a steroidal antiandrogen, decreased male territorial aggression in a lizard (Anolis carolinensis) (Deckel, 1996). In addition, flutamide, a nonsteroidal antiandrogen, was found to reduce aggressive behavior in the European robin (Erithacus rubecula) (Schwabl and Kriner, 1991) and red-winged blackbird (Agelaius phoeniceus) (Searcy and Wingfield, 1980). In order to further our understanding of androgenic modulation of aggressive behavior we are using male song sparrows, Melospiza melodia morphna, as a model system. We examined the potential role of AR in regulating territorial aggression in both freely living and captive male song sparrows using flutamide to block AR function. The endocrine control of aggression in song sparrows has been extensively studied with a particular emphasis upon the seasonal and environmental control of aggression as well as the hormonal correlates to these behaviors (Soma et al., 2000a,b; Wingfield, 1994; Wingfield and Hahn, 1994). These birds defend multi-purpose breeding and feeding territories year-round and possess a well-defined set of behaviors consisting of combinations of visual displays and vocalizations that are involved with territorial defense against conspecific intruders (Nice, 1943; Wingfield, 1985; Wingfield and Monk, 1992). These behaviors can be readily measured in freely living birds using a standardized simulated territorial intrusion (STI) (Wingfield, 1985) as well as in captive birds held in controlled environments using a laboratory simulated territorial intrusion (LSTI) (Sperry et al., 2003). Studying both freely living and captive song sparrows allows us to examine birds under conditions where the myriad of social and environmental cues are all present as well as look at the birds' behaviors on a finer scale in the laboratory. Within this paper we examine the role that ARs play in the control of aggression under both natural and laboratory conditions. Freely living birds were studied in the early part of the breeding season in order to focus on the time period when the male birds transition from their winter non-breeding life history stage to their spring breeding life history stage. During their non-breeding state, male song sparrows have regressed gonads and low to non-detectable circulating plasma T levels. As the birds move from this life history stage into the breeding life history stage, their levels of T increase as they establish territories and actively work to acquire and then defend mates and territories (Wingfield, 1985; Wingfield and Monk, 1992). Due to the rapid changes in levels of circulating T as the birds move between life history stages, we predicted that during these time periods the male song sparrows would be most sensitive to treatment with flutamide. For the purposes of this paper the term pre-breeding refers to the time period when male song sparrows are present on

87

their breeding territories, actively acquiring mates and possibly involved in pair bonding with potential mates but not yet ready to breed. Breeding refers to the time period early in the spring when the male birds have first begun to actively breed. Both are parts of the breeding life history stage with the pre- and early breeding time periods representing different substages. Note that song sparrows in the Pacific Northwest raise multiple broods and will breed throughout the late spring and summer. The laboratory studies examined both winter, non-breeding, aggression that is equivalent to short-day (SD) laboratory conditions and spring, breeding, aggression that is equivalent to long-day (LD) conditions. LD birds were also used to determine whether flutamide treatment had any unexpected behavioral or physiological side effects. In both cases, freely living song sparrows actively defend territories allowing us to address the question of whether ARs play a role in modulating breeding and non-breeding aggression. Methods Drugs Flutamide (α-trifluoro-2-methyl-4′-nitro-m-propionotoluidide) was purchased from Sigma-Aldrich Co (product F9397). This drug was administered using silastic implants placed subcutaneously on the flanks of each song sparrow. Field behavioral studies The field studies were conducted in Western Washington State, USA between March 9 and April 19, 2001 and between April 20 and May 24, 2000. These time periods represent the song sparrows prebreeding and early breeding life history substages, respectively (Wingfield and Hahn, 1994). Male song sparrows were caught using mist nets combined with conspecific playback. After capture, birds were immediately removed from the net and a blood sample was taken from the alar wing vein for determination of blood androgen levels. Measurements were taken for wing and tarsus length, mass and the length and width of the cloacal protuberance (CP). From the length and width measurements, a volume was calculated using the formula for a cylinder. In addition, abdominal and furcular fat was visually assessed using a standard five-point scale (see Wingfield and Farner, 1978). Birds were banded with fish and wildlife aluminum bands and given a unique set of color bands. In addition to body morphometrics, time to capture was determined. Time to capture is the amount of time elapsed between initiating song playback and the bird flying into the mist-net. After body morphometrics were recorded, each bird received three of either control (empty) or flutamide filled silastic implants. Silastic implants were 12 mm in length (1.47 mm i.d. and 1.96 mm o.d.) and sealed on both ends with 1 mm of clear silicone rubber sealant. Each implant was inserted through a small incision in the skin and sealed with nexaband liquid topical tissue adhesive (Abbott Laboratories, Chicago, IL, USA) followed by New-skin liquid bandage (Medtech, Chicago, IL, USA). Two implants were placed onto the left flank and the third implant was placed onto the right flank. Territorial aggression for each male was quantified using a simulated territorial intrusion (STI) (Wingfield, 1994) 18 days after the birds received silastic implants. STIs were conducted using a caged conspecific decoy placed near the center of the territory combined with conspecific playback. Care was taken to place the decoy away from the location where the focal bird was previously captured. Two different playback tapes were used and each contained songs from a single male repeated every 10 s with a 1 min silence after every 2 min of song. The song type changed after each 2 min of song. The STI consisted of 10 min of observation of the focal bird with the decoy present and playback on. The persistence portion of the STI included 10 min of observation to measure the persistence of the response with

88

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

no decoy present and no song playing. During the STI, behaviors were recorded continuously using a hand-held tape recorder. The number of songs, number of flights, closest approach, time spent within 5 m of the decoy and the latency to respond to the decoy and song were all recorded. Response latency was not measured as part of behavioral persistence. During the study equal numbers of birds were used at two different sites, Pack Forest, WA and the Skagit Valley Wildlife Reserve, WA. All of the song sparrows were on non-adjacent territories within each year and several of the same territories were used in both 2000 and 2001, but no individuals were used in multiple years. All experiments were conducted in accordance with University of Washington IACUC regulations. Laboratory behavioral studies Male song sparrows were captured using mist-nets and conspecific playback in September 2001 in Western Washington, USA. After capture, sparrows were housed in outdoor aviaries at the University of Washington where they experienced natural photoperiod and temperature. For behavioral studies, song sparrows were placed into individual wire cages (35 × 40 × 42 cm) and housed in environmental chambers at 15 °C where they were given ad libitum water, bird chow (Nutrition International, Brentwood, MO) and wild birdseed mix (University of Washington Custom Seed Mix, Seed Factory NW, Inc, Kent, WA) and provided with water baths. Grit fortified with vitamins and minerals were given weekly (Bird Health Grit, Seed Factory, Ceres, CA). Each song sparrow was visually isolated with curtains from its nearest neighbors. Two groups of song sparrows were used: short-day (SD) photoperiod birds (8L:16D) and long-day (LD) photoperiod (16L:8D) birds. Both sets of birds were implanted with either control (empty) or flutamide filled silastic implants under isoflurane anesthesia in an identical manner to the birds used in the field studies. Each bird received 3 implants, each 12 mm in length. The SD birds were placed into cages immediately after the insertion of the silastic implants and held 17–23 days prior to the start of behavioral testing (see below). The LD birds were placed into individual cages in late November, after they were naturally photo-sensitive, and the silastic implants were inserted. The birds were held on 16L:8D photo regime for 24 days prior to the start of behavioral observations in order to maximally stimulate the hypothalamo–pituitary–gonad (HPG) axis and allow the gonads to recrudesce (Wingfield, 1993). At the end of both SD and LD studies blood was collected from the alar wing vein from each bird for measurement of androgen levels.

started 18 days after the implants were placed into the birds, while the LD birds were tested 42 days after implants were placed into the birds. The researchers remained blind to the treatment groups until after the data were analyzed. The birds were transferred, in their home cage, to the testing chamber 1 h prior to behavioral testing. The focal bird's cage was placed on the left side of the decoy cage from which it was visually isolated with a curtain. Behavioral trials were observed from outside the chamber behind a one-way mirror and videotaped using a Sony digital camera (T700, Sony Corp, Park Ridge, NJ). The laboratory simulated territorial intrusion (LSTI) consisted of recording 10 min of behavior after the curtain was lifted and audio playback initiated. Five minutes of pre- and post-LSTI behavior was also recorded, but not used for behavioral analysis within this study. A decoy novel to the test subjects was used for each set of trials. The playback tape consisted of 2 min of conspecific song followed by 1 min of silence repeated for 10 min with a new set of songs each cycle. At the end of the trials, the latency to respond to the stimuli was recorded and behavior was quantified using hierarchical scales of direct and indirect aggression (see Sperry et al., 2003 for more details). Aggressive behavior was scored and ranked for each second of the 10 min LSTI. The scale for direct aggression includes (score): no aggressive posture (0), flattened posture with wings held at sides (1), wing tips slightly drooped (2), wing tips fully drooped (3), and wings flared out from side (4). A score of 0.5 was added for the combined behaviors of bill gape, tail flare and growl and 1 was added for a charge. The scale for indirect aggression includes: no aggressive posture (0), bill wipes (1), wing waves (2), song (3) and song plus wing waves (4). A score of 0.1 was added to each behavior when the bird combined the above indirect behaviors with feather puffing. Radioimmunoassay Blood samples, stored on ice when in the field, were spun in a centrifuge and the plasma separated and then frozen at − 20 °C until the hormone assays were completed. All blood samples were analyzed in duplicate by radioimmunoassay following the procedures of Wingfield et al. (1991). The samples were analyzed using a direct assay with no chromatography and thus measured levels of total androgen (testosterone + 5α-dihydrotestosterone). The assay detection limit ranged from 0.06 to 0.19 ng/ml and was dependent upon the individual sample volume. Separate assays were run for the samples collected in the two field-seasons and for the LD and SD laboratory-held birds.

Laboratory behavioral studies: Activity time budgets

Statistics

To determine whether flutamide implants had an effect upon general activity, activity scans were conducted on LD birds 18 days after receiving implants. Scans were conducted during the first 3 h after lights on for three consecutive days starting 15 min after lights on in the morning and lasting for 40 min, followed by a 20 min break (see Martin and Bateson, 1998). The birds were observed in their individual cages within the environmental chambers from behind a one-way glass mirror. During each minute an instantaneous determination of each individual's behavior was made, thus each bird received 40 such determinations per hour. The first 3 h after the lights come on in the chamber encompasses the most active period for the birds (Sperry et al., 2003). Five different categories of behaviors were measured: inactive (standing, resting and puffed), active (hopping, moving along perch), grooming (preening and bathing), feeding (eating at feeder, foraging on bottom of cage and drinking) and singing.

Activity data from LD laboratory song sparrows were averaged for each bird across the three time blocks and the 3 days to provide an average percentage for each behavior. These behaviors were statistically analyzed for the effect of flutamide treatment with a one-factor MANOVA using JMP statistical software (SAS Institute, NC USA). Laboratory behavioral data were analyzed with JMP using either a Wilcoxon rank sum test or a t-test for the combined direct and indirect behavioral scores and the response latency, respectively. Field behavioral data were analyzed with t-tests with Bonferroni corrections for use of multiple tests. The song sparrow morphometric data and testosterone data were analyzed using t-tests and Wilcoxon rank sum tests for ordinal data and a MANOVA for LD laboratory birds. Data are presented as means ± SEM.

Laboratory behavioral studies: Measurement of aggressive behavior

Field morphological data

Trials for both the SD and LD birds were performed in an identical manner with 5 trials run per day beginning at 0900. SD birds were

Prior to the start of the pre-breeding study the two treatment groups of song sparrows did not significantly differ with regard to any

Results

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

measurement of body morphometrics. After the behavioral trials ended, the song sparrows were recaptured to determine whether the implants altered body morphometrics. For the pre-breeding song sparrows 7 of 8 control and 5 of 9 flutamide implanted birds were recovered. Wing length ranged from 64.8 ± 0.7 mm in control birds (n = 10) compared to 65.2 ± 0.9 mm in flutamide birds (n = 10) (t18 = −0.12, p = 0.91). For control birds, weight was 26.5 ± 0.4 g compared to flutamide birds at 26.5 ± 0.4 g (t17 = 0.29, p = 0.77). After the end of each study, measurements were repeated to assess the effects of treatment. The data were analyzed as the change after treatment and the statistics are for differences between each treatment population. Weight did not significantly differ between treatment groups. Control birds (n = 7) changed 0.4 ± 0.4 g over the course of the study compared to flutamide treated birds (n = 5) which changed 1.2 ± 0.8 g (t10 = − 0.9, p = 0.35). The initial fat score for control birds was 0.85 ± 0.1 with a change of 0.1 ± 0.2 and in flutamide birds the starting fat score was 0.67 ± 0.2 with a change of 1.0 ± 0.4 (initial: z = −1.1, p = 0.28; change: t10 = −2.08, p = 0.064). Initial CP volume in control birds was 91.6 ± 22.2 mm3 compared to 90.2 ± 18.5 mm3 in flutamide birds (t18 = 0.050, p = 0.75). In order to assess the effect of treatment time, CP volume is expressed as a growth rate to account for the varying number of days each bird was implanted. In control birds the CP grew at 9.1 ± 2.0 mm3/day compared to 10.1 ± 2.5 mm3/day (t10 = −0.33, p = 0.75). Like the pre-breeding cohort of birds, the birds used in the breeding substage study did not differ with regard to any morphometric measurement. For the song sparrows in the group of breeding birds 3 of 7 control and 5 of 8 flutamide implanted birds were recovered. In control birds (n = 9), wing length measured 64.7 ± 0.9 mm compared to 65.7 ± 0.6 mm in flutamide birds (n = 8) (t15 = − 0.91, p = 0.38). For control birds weight was 25.6 ± 0.5 g compared to 26.8 ± 0.4 g in flutamide birds (t15 = −1.9, p = 0.076). The change in weight after treatment did not significantly differ between treatment groups. Control birds (n = 3) decreased −0.7 ± 0.3 g over the course of the study compared to flutamide treated birds (n = 5) which increased 0.1 ± 0.3 g (t6 = −1.7, p = 0.14). The initial fat score for control birds was 0.6 ± 0.1 with a change of 0.33 ± 0.3 and in flutamide birds the starting fat score was 0.6 ± 0.04 with a change of 0.1 ± 0.25 (initial: z = 0.054, p = 0.96; change: t6 = 0.59, p = 0.58). Initial CP volume in control birds was 444.8 ± 37.3 mm3 with a negative rate of growth at −6.0 ± 1.6 mm3/day compared to 440.0 ± 27.3 mm3 and a negative rate of growth of −1.4 ± 2.2 mm3/day in flutamide birds (initial: t15 = 0.18, p = 0.86; change: t6 = −1.8, p = 0.12). In addition to body morphometrics, the time to capture each song sparrow, via mist-net, was measured. The initial capture was accomplished with playback only, while the recapture at the end of the study was done with the addition of the conspecific decoy. For the pre-breeding song sparrows there were no differences in the time to capture, taking 5.7 ± 1.1 min to capture each control bird compared to 3.5 min for the treatment birds (t18 = 0.21, p = 0.83). At the end of the study, there was a significant increase in the amount of time it took to recapture the birds implanted with flutamide compared to those with empty implants. The flutamide birds took an additional 24.1 ± 5.0 min compared to an increase in 11.8 ± 2.9 min with the control birds (t10 = − 2.27, p = 0.046). During the breeding substage there were no differences in capture times either before the start or at the end of the study. The control birds took 8.0 ± 3.5 min to capture at the start of the study with an increase of 8.2 ± 6.5 min at the end of the study. Similarly, the flutamide treated birds took 5.2 ± 2.1 min to capture and an additional 17.6 ± 5.4 to recapture (initial: t15 = 0.67, p = 0.51; change: t6 = − 1.09, p = 0.32). Field behavioral data During the pre-breeding substage, 8 of the 10 song sparrows that received control implants and 9 of the 10 birds that received flutamide

89

implants remained on their territories 3 weeks later when the STIs were conducted. For the 3 song sparrows that were not found on their territories non-banded male song sparrows were present and found to be defending the territory presumably replacing the previously banded and implanted male. The pre-breeding song sparrows were implanted with empty implants for 20.1 ± 2.3 days or flutamide implants for 23.3 ± 1.9 days. During the breeding substage, 7 of the 9 control implanted and 8 of 8 flutamide implanted song sparrows remained on their territories 3 weeks later when the STIs were conducted. The breeding birds received empty and flutamide filled implants for 18.9 ± 0.9 and 18.6 ± 0.6 days respectively before the STIs were completed. There was no change in plasma androgen levels as a result of flutamide implants during either the pre-breeding or breeding substage. During the pre-breeding substage, pre-implant androgen levels were 2.4 ± 0.80 ng/ml (n = 8) in control birds compared to 3.1 ± 0.65 ng/ml (n = 7) in treatment birds (t13 = 0.94, p = 0.36). In these same birds, post-implant levels were 2.4 ± 0.96 ng/ ml (n = 9) in control birds compared to 3.8 ± 2.5 ng/ml (n = 4) in treatment birds (t11 = 0.57, p = 0.58). During the breeding substage, post-implant plasma levels in control birds were 2.1 ± 1.1 ng/ml (n = 3) compared to treatment birds at 0.63 ± 0.14 ng/ml (n = 5) (z = 1.5, p = 0.14). Five measures of aggression were quantified during the 10 min STI and statistically analyzed using individual t-tests with a Bonferroni correction for the use of multiple statistical tests (α = 0.01). During the pre-breeding substage, the total number of aggressive flights was significantly lower in the song sparrows implanted with flutamide (6.4 ± 1.5) compared to those with empty implants (16.0 ± 1.6) (Fig. 1A). However, during the breeding substage there was no difference between birds with control (3.9 ± 1.8) or flutamide (3.6 ± 1.6) implants (pre-breeding: t15 = 3.60, p = 0.0026; breeding: t13 = 0.13, p = 0.90). There were no significant differences between treatment groups for songs during the pre-breeding substage 32.2 ± 5.3 in control birds and 24.1 ± 5.0 in flutamide birds (t15 = 1.07, p = 0.30) (Fig. 1B). Similar results were seen during the breeding substage with controls at 24.1 ± 5.7 songs compared to 15.4 ± 5.3 songs for flutamide treated birds (t13 = 1.21, p = 0.25) (Fig. 1B). During the pre-breeding substage song sparrows showed a closest approach to the decoy of 2.1 ± 0.96 m compared to 2.9 ± 0.85 m in control and flutamide treated birds respectively (t15 = − 0.81, p = 0.43) (Fig. 1C). During the breeding substage the birds had a closest approach of 4.0 ± 0.96 m and 3.7 ± 0.90 m in control and flutamide treated birds respectively (t13 = 0.21, p = 0.84) (Fig. 1C). The pre-breeding control song sparrows remained within 5 m of the decoy during the STIs for 281.5 ± 56.8 s compared to flutamide birds which spent 236.3 ± 54.5 s (t15 = −0.64, p = 0.53) (Fig. 1D). The breeding birds spent slightly less time within 5 m of the decoy with control birds at 195.3 ± 60.7 s compared to flutamide birds at 133.6 ± 56.8 s (t13 = 0.68, p = 0.51) (Fig. 1D). Finally, the latency to respond to the STI was not different for control birds at 149 ± 51.9 s and flutamide birds at 157 ± 61.1 s (t15 = − 0.12, p = 0.91) (Fig. 1E). In breeding birds the response was similar with control song sparrows at 138.0 ± 55.5 s and flutamide birds at 116.0 ± 51.9 s (t13 = 0.29, p = 0.78) (Fig. 1E). During the persistence portion of the STI song sparrows showed no significant differences as a result of treatment with flutamide during either the pre- or the breeding substages (Table 1). Short-day laboratory study Song sparrows implanted with flutamide (n = 8) for 18–24 days were no more aggressive than birds implanted with empty silastic implants (control, n = 8). Total aggression scores were 36.3 ± 14.6 and 65.3 ± 37.2 for control and flutamide treated birds, respectively (z = − 0.105, p = 0.916). The latency to the expression of first aggressive behavior for control birds (88.1 ± 46.1 s) did not differ

90

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

Fig. 1. Six behavioral responses associated with territorial aggression during simulated territorial intrusions in freely living, male song sparrows implanted with either flutamide or blank implants (control) during pre-breeding and breeding substages. Flutamide significantly decreased aggressive flights during the pre-breeding substage, but had no effect during the breeding substage when aggressive flights are naturally fewer in number (A). Flutamide did not reduce the other four measures of aggression (B–E). Statistics are for differences between each treatment group within each substage with α = 0.01 for significant differences.

from flutamide treated birds (91.0 ± 54.1) (z = 0.052, p = 0.958). There were no differences in wing length between the group of control birds, 65.1 ± 0.8 mm, compared to the flutamide group, 66.1 ± 0.5 (t14 = −1.06, p = 0.31). Pre-implant body weight of the control group was 23.5 ± 0.5 g and changed 0.7 ± 0.5 g over the course of the study. This was not significantly different from the flutamide group which weighed 22.8 ± 0.6 g pre-implant and changed 0.3 ± 0.9 g (initial: t14 = 0.90, p = 0.38; change: t14 = 0.32, p = 0.75). Note that the statistics are for the effect of treatment on changes in morphometrics. Likewise, the fat score

was not significantly different pre-implant between the control group (0.6 ± 0.4) and the flutamide group (1.0 ± 0.3) (z = 0.93, p = 0.35). Neither did the treatment significantly effect fat score since control birds increased fat by 1.4 ± 0.4 and flutamide birds increased by 0.3 ± 0.3 (z = −1.74, p = 0.081). In addition, there was no change in plasma androgen levels as a result of flutamide implants during the short-day laboratory study. The post-implant plasma levels in control birds were 0.15 ± 0.04 ng/ml (n = 8) compared to treatment birds at 0.13 ± 0.02 ng/ml (n = 8) (t14 = 0.63, p = 0.53).

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

91

Table 1 Mean aggression scores for the four measures of persistence of aggression in the field.

Number of flights Number of songs Closest approach (m) Time within 5 m (s)

Treatment

Pre-breeding

Control Flutamide Control Flutamide Control Flutamide Control Flutamide

9.1 ± 1.3 6.8 ± 2.4 22.9 ± 8.2 16.6 ± 4.9 5.0 ± 1.2 6.1 ± 1.6 156.0 ± 73.4 139.2 ± 64.1

Breeding t15 = 0.84, p = 0.42 t15 = 0.68, p = 0.51 t15 = − 0.56, p = 0.58 t15 = 0.17, p = 0.86

3.9 ± 2.6 3.4 ± 2.8 21.3 ± 8.5 21.1 ± 5.5 5.6 ± 1.9 6.3 ± 2.3 90.7 ± 45.3 82.1 ± 45.5

t13 = 0.25, p = 0.80 t13 = 0.016, p = 0.99 t13 = − 0.27, p = 0.79 t13 = 0.13, p = 0.90

Statistics are for differences between each treatment group within each substage.

Long-day laboratory study Flutamide treatment resulted in a significant decrease in aggressive behavior of male song sparrows held on long-day photoperiod (z = −1.96, p b 0.0489) (Fig. 2A). However, there was no difference between treatment groups in the latency to respond to the male song sparrow decoy (t13 = − 0.46, p b 0.65) (Fig. 2B). The general activity levels of male song sparrows were examined in birds held on long-day photoperiod (16L:8D) for 18 days and treated with either flutamide implants (n = 8) or empty silastic implants (n = 7) (Fig. 3). The data were placed into a time budget and analyzed using a MANOVA. There was no treatment effect (F(5,9) = 0.95, p b 0.228) demonstrating that flutamide does not change the activity levels of male song sparrows. Over the course of the 42 days of implants there were few effects of flutamide on body morphometrics. There was a significant interaction between time and treatment on the weight, but that was due to a difference in weight of the two groups of birds prior to placing the implants into the song sparrows. Control birds weighed 25.1 ± 0.5 g at the start of the study and changed −0.1 ± 0.7 g and −0.9 ± 0.6 g after 18 and 42 days, respectively. Flutamide birds initially weighed 23.6 ± 0.3 g and after 18 days changed by −0.1 ± 0.7 g with a final 42 day weight change of 1.0 ± 0.3 g (treatment: F(1,13) = 0.06, p = 0.386; time: F(2,12) = 0.923, p = 0.423; time ⁎ treatment: F(2,12) = 3.89, p = 0.0498). The fat score did not differ between treatment groups or over time during this study. Control birds had an initial score of 2.9 ± 0.1, an 18 day decrease of 0.6 ± 0.3 followed by a day 42 decrease of 0.5 ± 0.2. The flutamide group started with a score of 2.4 ± 0.2 with an 18 day decrease of 0.1 ± 0.4 and no additional change at day 42 with a score of 0.0 ± 0.3 (treatment: F(1,13) = 0.91, p = 0.358; time: F(2,12) = 1.37, p = 0.291; time ⁎ treatment: F(2,12) = 0.442, p = 0.442). There was

Fig. 2. The general activity levels of male song sparrows held on long days (16L:8D) and implanted with either flutamide (n = 7) or empty silastic implants (n = 8) for 18 days. Data were placed into a time budget and analyzed using a one-factor MANOVA where no differences in activity levels were found.

a significant increase in the size of the CP over time, but no effect of treatment. Thus, this is the expected growth of the CP as the birds undergo their seasonal reproductive cycle. Initially, pre-implant CP volume in control birds was 37.1± 6.5 mm3 followed by increases at day 18 of 34.3 ± 9.2 mm3 and at day 42 of 138.4 ± 22.2 mm3. Within the flutamide group, CP volume was initially 40.4 ± 4.7 mm3 with an increase at day 18 of 21.3 ± 10.5 mm3 and an increase by day 42 of 129.1 ± 42.8 mm3 (treatment: F(1,13) = 0.0585, p = 0.813; time: F(2,12) = 19.9, p = 0.002; time ⁎ treatment: F(2,12) = 0.40, p = 0.68). There was no change in plasma androgen levels as a result of flutamide implants over the duration of this study. At 3 weeks postimplant, plasma levels in control birds were 0.86 ± 0.27 ng/ml (n = 8) compared to treatment birds at 0.71 ± 0.13 ng/ml (n = 7) (t13 = 0.50, p = 0.62). Similarly, at 6 weeks post-implant plasma levels in control birds were 1.11 ± 0.33 ng/ml (n = 8) compared to treatment birds at 0.76 ± 0.08 ng/ml (n = 7) (t13 = 0.99, p = 0.33).

Fig. 3. Behavioral responses in male song sparrows during a laboratory simulated territorial intrusion. The cumulative aggression score (A) and the latency to respond (B) after exposure to a conspecific decoy and audio playback. Birds were treated with either flutamide implants (n = 7) or empty silastic implants (control) (n = 8) for 42 days.

92

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

Discussion Experimental treatment of male song sparrows with the antiandrogen flutamide decreased some aggressive behaviors during the pre-breeding substage (March–April) in freely living birds as well as in laboratory-held birds under LD (16L:8D) conditions. These results suggest that the AR plays an important regulatory role in the control of male territorial aggression in song sparrows. Interestingly, during the breeding substage, the end of April through the end of May, there was no measurable effect of flutamide treatment on aggressive behavior. Thus, this suggests that in song sparrows AR dependent control of aggressive behavior changes as life history states change. Freely living birds Treatment with the AR antagonist flutamide produced a modest decrease in aggressive behavior as indicated by a reduction in the number of flights associated with aggressive display during the prebreeding substage. Qualitatively, these are flights associated with the aggressive response to a territorial intrusion that range from short flutter flights between bushes to longer flights across the territory. The CP volumes of the pre-breeding birds were small and growing during the time period that these birds were implanted with flutamide. This indicates that these birds were implanted prior to the initiation of breeding. This is in contrast to the fully developed CPs found in the breeding birds. Interestingly, the aggressive flights were also reduced in both the control and flutamide treated birds in the breeding substage. During the STI, the birds with a reduced number of flights tended to approach the decoy from within bushes, staying away from exposed perches, rather than the frequent flights between bushes and across the territory. In addition to the aggressive flights that were directly quantified during the STI, the differences in recapture time after the STI ended provide a second, but indirect, measure of this behavioral change. Each song sparrow was captured with mist-nets at the start of each study using only playback and a second time was attempted after the STI ended using both playback and a decoy. However, not all of these second attempts were successful. In both cases, capturing these birds takes advantage of the increase in aggressive flights in response to the external stimuli. Thus, the increase in time required to recapture the flutamide treated song sparrows compared to the controls correlates with the drug-induced decrease in aggressive flights. There was not a significant difference between treatments in the breeding substage, however less than half of the control birds were recaptured at the end of the study. This result is consistent with the males mate-guarding and not actively flying around their territories. Based upon the known life history traits for the song sparrow (Wingfied and Monk, 1992; Wingfield and Hahn, 1994), the breeding time period corresponds to the initiation of egg-laying by the female and mate-guarding by the male. The decreased flights are consistent with mate-guarding behavior in the song sparrow as well as other monogamous, territorial passerines. During mate-guarding, males are much less likely to venture away from the fertile female in order to minimize potential extra-pair copulations by intruding males. Thus, the aggressive response of the birds treated with flutamide to a territorial intrusion during the pre-breeding substage appears to be qualitatively as well as quantitatively similar to the response of control song sparrows during the breeding substage when they are mate-guarding. These results indicate that this shift in behavior from long, active fights during territorial defense to fewer, shorter flights is under AR regulation. Given the dynamic nature of aggressive behavior (see Wingfield et al., 2006) as the birds move through the various life history stages and substages it is not surprising that AR regulation is also dynamic. These results suggest that these specific aggressive flights may be under AR regulation. However, these data do not provide any specific

insight as to the mechanism through which ARs may be regulating flights. For example, ARs could directly regulate muscle motor-control or instead regulate a motivational aspect of aggressive flights. In male golden-collared manakins (Manacus vitellinus), the spinal cord neurons controlling the flight muscles that are used during the highly choreographed male courtship displays possess putative ARs (Schultz and Schlinger, 1999). The authors (Schultz and Schlinger, 1999) suggested that androgens may influence the neuromuscular control of wing movement and potentially help coordinate the complex patterns of manakin courtship behaviors. Similar to our study, Schwabl and Kriner (1991) found that treatment of European robins with flutamide did not affect song rates, closest approach or time within 5 m during an STI. However, they did find that flutamide significantly increased the latency to respond to the initial stimulus. It is possible that similar mechanisms are responsible for the decrease in flights in song sparrow and response latency in European robins. However, it is key that neither the Schwabl study nor this study could find a decrease in overall aggression with regard to male–male territorial behavior. Laboratory-held song sparrows One important question that arose from the field studies concerned whether the decrease in flights was specific to the aggressive response or alternatively whether there was a more general decrease in overall activity. The activity data generated from the LD laboratory-held song sparrows demonstrated no difference in activity levels after flutamide treatment. Thus, we can reasonably conclude that the decrease in flights is likely not due to a generalized effect on activity. Further support for this conclusion comes from our field data. There was only one song sparrow not found on its original territory after treatment with flutamide, suggesting that overall activity levels likely were not dramatically altered. A general decrease in the ability to fly would presumably have a negative effect on the survival of freely living song sparrows. We examined the role of ARs in regulating aggressive behavior in both SD and LD male song sparrows. The results demonstrate that aggression is reduced in LD birds, but not in SD birds. However, in the laboratory SD birds demonstrated low levels of aggression, unlike what is seen in freely living birds during the non-breeding season where their levels of aggression are similar to breeding birds (Wingfield and Monk, 1992; Wingfield and Hahn, 1994; Soma et al., 1999, 2000a). Because there were relatively low levels of aggression under SD conditions it was not possible to demonstrate a decrease in behavior. This suggests that the SD results should be interpreted carefully and they probably do not demonstrate a lack of AR dependent regulation of aggression in the SD birds. Interestingly, non-breeding aggression in freely living song sparrows has been shown, at least in part, to be controlled by the low levels of circulating testosterone that could be blocked by treatment with aromatase inhibitors (Soma et al., 1999, 2000a). These studies clearly show that steroids, and particularly E2 is an important regulator of non-breeding territorial aggression. The role for the androgenic component of T remains to be determined. Unlike the SD birds, the LD birds showed a significant reduction in overall levels of aggression. The decrease in aggression was quantitatively different from what was measured in freely living birds. The reduction in aggression was not isolated to one specific behavior rather there was a reduction in multiple components of the suite of aggressive responses. Importantly, there was also a decrease in songs in the LD birds. The lack of response of SD birds compared to LD birds suggests that under laboratory conditions the cues necessary to elicit non-breeding aggression may be different from the cues that elicit LD, or breeding, aggression. This is not completely unexpected as the non-breeding aggression in song sparrows is thought to occur within the context aggression over food resources rather than sexual resources (Wingfield, 1985; Wingfield and Monk, 1992).

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

A. comparison of field and laboratory studies While a decrease in aggressive behavior in response to flutamide treatment was seen in LD song sparrows and pre-breeding song sparrows, the effect was much larger in laboratory-held birds. The fact that the responses to flutamide are different raise several interesting questions about the efficacy of flutamide as an antiandrogen and the transferability of laboratory results to in situ field studies. There are at least two plausible explanations that can account for the results—the differential efficacy of flutamide and/or the variations in behavioral assays. First, the levels of endogenous T may play an important role in determining the efficacy of flutamide. Since flutamide is a competitive inhibitor of AR, lower levels of endogenous androgens will increase the number of ARs inhibited by the antiandrogen. The levels of circulating androgens (testosterone and 5α-dihydrotestosterone) are lower in the laboratory compared to freely living song sparrows during the breeding season. As a result, it is likely that flutamide is a more effective antiandrogen in laboratory-held song sparrows compared to breeding freely living song sparrows. This leads to the notion that with a greater percentage of ARs blocked in laboratory-held birds, a broader suite of aggressive behaviors could be affected by the antiandrogen. The potential for such different effects suggests several important implications for this study, as well as other studies utilizing pharmacological blockade of receptors. It is inherently difficult to quantitatively determine the degree to which steroid receptors are inhibited in studies utilizing a pharmacological blockade, especially in studies where tissues are not harvested. Thus, when interpreting results from these types of studies, whether positive or negative, the results are indicative of only a subset of actual steroid targets. Furthermore, different steroid sensitive targets, whether behavioral or physiological, are not equally sensitive to pharmacological blockade. If the differential effects between laboratory and field results are due to the circulating androgen levels, then it indicates that the androgenic regulation of the flights is more sensitive to flutamide blockade than the androgenic regulation of the suite of behaviors measured in the laboratory. It is also likely that we are measuring only the most sensitive targets and there may well be more androgenic regulation of aggressive behavior that would be uncovered with more potent antiandrogens. A second approach to reconcile the laboratory and field results concerns the manner in which aggression was measured in the field and laboratory behavioral paradigms. While both measure aggressive behavior, the behaviors that are quantified are different subsets of the overall suite of aggressive behaviors expressed by song sparrows (Nice, 1943; Sperry et al., 2003; Wingfield, 1985). However, it is not known whether the laboratory model measures exclusively territorial aggression, or a more general male aggressive response. Thus, there are two scenarios that potentially account for the differences seen in the effect of flutamide on aggression between LD laboratory-held birds and the pre-breeding freely living birds. First, if both paradigms are eliciting a territorial response from birds, the effect of flutamide may simply be differentially apparent within the two paradigms. In other words, if the measured behaviors are along a continuum of a territorial response, the field results demonstrate the AR dependent behaviors on a large scale, whereas the laboratory results illuminate the fine-scale AR dependent behaviors. Alternatively, the neuroendocrine pathways controlling territorial versus general male–male aggression are differentially regulated by ARs. One argument in favor of the latter scenario is the different response to song in the field versus laboratory models. In the field, song was not reduced in the flutamide treated birds, however there was a large decrease in song rate in the flutamide treated LD birds. However, these two explanations are not mutually exclusive, and without further knowledge of the specific pathways of AR action or without a more complete

93

understanding of the specific actions of flutamide it is not possible to tease apart these two explanations. Few studies have compared laboratory versus field models of aggressive behaviors. Previous studies examining the serotonergic regulation of aggressive behavior have shown similar results between laboratory-held song sparrows (Sperry et al., 2003) and freely living American tree sparrows (Spizella arborea) (Sperry et al., 2005) albeit with fluctuating sensitivity in the freely living birds. Regardless of the potential differences between behaviors in the laboratory and field, this set of studies demonstrates that there is a certain subset of the suite of aggressive behaviors expressed by song sparrows that are under AR control. Flutamide—an AR antagonist? The effect of flutamide on aggression and other social behaviors in birds and in other vertebrates is varied. In other avian species, flutamide has been shown to decrease response latency in European robins (Schwabl and Kriner, 1991) and dominance in red-winged blackbirds (Searcy and Wingfield, 1980). Flutamide has also been an effective modulator of parental behavior in house sparrows (Passer domesticus) (Hegner and Wingfield, 1987) and song rates in blue-headed vireo (Vireo solitarus) (Van Roo, 2004) indicating that, in birds, it can effectively modulate putative AR dependent behaviors. However, in male golden-collared manakins T dependent courtship behavior was initially reduced and subsequently increased by flutamide treatment over the course of a 3 week study (Fusani et al., 2007). While the mechanism for this reversal is not known, it points to the degree of uncertainty that exists with regard to flutamide, or any drug, in terms of unexpected mechanisms of actions or side effects. In addition to studies of behavior, flutamide has been used in birds to examine the role of AR in the development of secondary sexual characters (e.g. Wade and Buhlman, 2000) as well as androgen regulation of the AR (Nastiuk and Clayton, 1994). Interestingly, in other classes of vertebrates flutamide has been ineffective in modulating aggressive behavior. This is illustrated in the lizard Anolis carolenensis where cyproterone acetate, a steroidbased antiandrogen, reduces male territorial aggression whereas flutamide does not (Tokarz, 1987). In this study, flutamide increased testosterone levels as a result of decreased GnRH negative feedback, thus likely limiting its effectiveness as an antiandrogen (Tokarz, 1987). Similarly, in mammals flutamide has not been effective in decreasing aggression in intact rats, mice or hamsters (e.g. Heilman et al., 1976). However, it has been shown to effectively decrease sexual behaviors in mice only after castration (Vagell and McGinnis, 1998). In mammals, flutamide has been shown to increase levels of androgens via decreased negative feedback (Mahler et al., 1998; Ohsako et al., 2003). The addition of a GnRH analogue increases the efficacy of flutamide in the treatment of androgen dependent cancers (Mahler et al., 1998). This difference between the effectiveness of flutamide as an antiandrogen between birds and other classes of vertebrates highlights the critical importance of androgenic plasma steroid levels in determining the effectiveness of flutamide as an antiandrogen. One of the differences between birds and mammals and likely other vertebrates is that in birds, flutamide does not appear to disrupt the negative feedback of T. Neither plasma LH nor T increases after treatment with flutamide in house sparrows (Hegner and Wingfield, 1987). In birds, it appears that estrogen, via the ER, controls androgen levels via negative feedback (Soma et al., 2000a). The fact that androgen levels remain at normal physiological levels during treatment with flutamide should increase the likelihood that flutamide will effectively inhibit androgen action via ARs. This strongly suggests that lower androgen levels may explain why flutamide is more effective as an antiandrogen in birds than in other

94

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95

vertebrates. This further illustrates the importance that the levels of endogenous androgens play in determining the effectiveness of flutamide as an antiandrogen. In male passerine birds, the CP is an androgen dependent, sperm storage organ that is frequently used as a morphological indicator of the presence of high circulating plasma androgen levels. Interestingly, but not unexpectedly, the growth of the CP during the breeding season was not reduced by flutamide. This lack of effect of the antiandrogen has been demonstrated previously in male house sparrows (Hegner and Wingfield, 1987). Thus, it is likely that this androgen-dependent organ either requires only low levels of T for maximum growth, or is relatively insensitive to flutamide treatment. This is a further indication that while this study does demonstrate an important role for the AR in regulating aggressive behavior there are clearly androgen-dependent events that are not reduced with flutamide treatment. Thus, it is possible that there are additional androgen-dependent aggressive behaviors that are not blocked by this treatment. Conclusions This study shows that competitively inhibiting AR function with the antiandrogen flutamide reduces male–male aggressive behavior in both freely living and laboratory-held song sparrows. This demonstrates that ARs are an important component of the neuroendocrine control of aggressive behavior and lends additional support to work previously done in other vertebrates (e.g. Sato et al., 2004; Deckel, 1996; Schwabl and Kriner, 1991; Searcy and Wingfield, 1980). However, given the limitations in effectively measuring AR function it is also likely that neither this study nor previous studies are measuring the full extent of AR involvement in modulating male aggression. Future studies examining specific AR dependent neuroendocrine pathways are needed to begin to tease apart AR function and its role in regulating male aggression. One additional intriguing possibility is the currently unknown role that the multiple ARs have in modulating behaviors such as aggression. While multiple ARs have not been found in all vertebrates, a variety of multiple isoforms and/ or subtypes have been found in frogs (Fischer et al., 1993), fish (Ikeuchi et al., 1999; Sperry and Thomas, 1999) and mammals (Wilson and McPhaul, 1996). The potential for tissue and cell specific responses due to AR multiplicity may prove to be important pieces for a more complete understanding of AR function in mediating aggressive behavior. Acknowledgments This research was supported in part through the NSF grant IBN9905679 to J.C.W., and through the University of Washington, NIH Reproductive Biology Training Grant as well as the NIMH NRSA grant MH12986-01 to T.S.S. We would like to thank Chanira Reang Sperry, Haruka Wada, and Latisha Wright for field assistance. Lynn Erckmann provided valuable assistance with animal care. In addition, we thank two anonymous reviewers for offering helpful suggestions and edits. References Archawaranon, M., Wiley, R.H., 1988. Control of aggression and dominance in whitethroated sparrows by testosterone and its metabolites. Horm. Behav. 22, 497–517. Canoine, V., Gwinner, E., 2002. Seasonal differences in the hormonal control of territorial aggression in free-living European stonechats. Horm. Behav. 41, 1–8. Deckel, A.W., 1996. Behavioral changes in Anolis carolinensis following injection with fluoxetine. Behav. Brain Res. 78, 175–182. Finney, H.C., Erpino, M.J., 1976. Synergistic effect of estradiol benzoate and dihydrotestosterone on aggression in mice. Horm. Behav. 7, 391–400. Fischer, L., Catz, D., Kelley, D., 1993. An androgen receptor mRNA isoform associated with hormone-induced cell proliferation. Proc. Natl. Acad. Sci. 90, 8254–8258. Fusani, L., Day, L.B., Canoine, V., Reinemann, D., Hernandez, E., Schlinger, B.A., 2007. Androgen and the elaborate courtship behavior of a tropical lekking bird. Horm. Behav. 51, 62–68.

Hegner, R.E., Wingfield, J.C., 1987. Effects of experimental manipulation of testosterone levels on parental investment and breeding success in male house sparrows. Auk 104, 462–469. Heilman, R.D., Brugmans, M., Greenslade, F.C., Davanzo, J.P., 1976. Resistance of androgen-mediated aggressive-behavior in mice to flutamide, an antiandrogen. Psychopharmacology 47, 75–80. Ikeuchi, T., Todo, T., Kobayashi, T., Nagahama, Y., 1999. cDNA cloning of a novel androgen receptor subtype. J. Biol. Chem. 274, 25205–25209. Mahler, C., Verhelst, J., Denis, L., 1998. Clinical pharmacokinetics of the antiandrogens and their efficacy in prostate cancer. Clin. Pharmacokinet. 34, 405–417. Martin, P., Bateson, P., 1998. Measuring Behaviour: An Introductory Guide, 2nd ed. Cambridge University Press, Cambridge. Nastiuk, K.L., Clayton, D.F., 1994. Seasonal and tissue-specific regulation of canary androgen receptor messenger ribonucleic acid. Endocrinology 134, 640–649. Nelson, R.J., 1995. An Introduction to Behavioral Endocrinology. Sinauer Assoc., Inc., Sunderland, MA. Nice, M. M. 1943. Studies in the Life History of the Song Sparrow. Volume II: The Behavior of the Song Sparrow and Other Passerines, Vol. 2. Dover, New York. Nomura, M., Durbak, L., Chan, J., Smithies, O., Gustafsson, J.-Å., Korach, K.S., Pfaff, D.W., Ogowa, S., 2002. Genotype/age interactions on aggressive behavior in gonadally intact estrogen receptor b knockout (bERKO) male mice. Horm. Behav. 41, 288–296. Ogawa, S., Robbins, A., Kumar, N., Pfaff, D.W., Sundaram, K., Bardin, C.W., 1996. Effects of testosterone and 7a-methyl-19-nortestosterone (MENT) on sexual and aggressive behaviors of two inbred strains of male mice. Horm. Behav. 30, 74-4. Ogawa, S., Washburn, T.F., Taylor, J., Lubahn, D.B., Korach, K.S., Pfaff, D.W., 1998. Modifications of testosterone-dependent behaviors by estrogen receptor-a gene disruption in male mice. Endocrinology 139, 5058–5069. Ohsako, S., Kubota, K., Kurosawa, S., Takeda, K., Qing, W., Ishimura, R., Tohyama, C., 2003. Alterations in gene expression in adult male rat testis and pituitary shortly after subacute administration of the antiandrogen flutamide. J. Reprod. Dev. 49 (4), 2003. Sato, T., Matsumoto, T., Kawano, H., Watanabe, T., Uematsu, Y., Sekine, K., Fukuda, T., Aihara, K., Krust, A., Yamada, T., Nakamichi, Y., Yamamoto, Y., Nakamura, T., Yoshimura, K., Yoshizawa, T., Metzger, D., Chambon, P., Kato, S., 2004. Brain masculinization requires androgen receptor function. Proc. Natl. Acad. Sci. U. S. A. 101, 1673–1678. Schlinger, B.A., Callard, G.V., 1989. Aromatase activity in quail brain: correlation with aggressiveness. Endocrinology 124, 437–443. Schultz, J.D., Schlinger, B.A., 1999. Widespread accumulation of [3H]testosterone in the spinal cord of a wild bird with an elaborate courtship display. Proc. Natl. Acad. Sci. U. S. A. 96, 10428–10432. Schwabl, H., Kriner, E., 1991. Territorial aggression and song of male European robins (Erithacus rubecula) in autumn and spring: effects of antiandrogen treatment. Horm. Behav. 25, 180–194. Searcy, W.A., Wingfield, J.C., 1980. The effects of androgen and antiandrogen on dominance and aggressiveness in male red-winged blackbirds. Horm. Behav. 14, 126–135. Simon, N.G., 2002. Hormonal processes in the development and expression of aggressive behavior. In: Pfaff, D.W., Arnold, A.P., Etgen, A.M., Fahrbach, S.E., Rubin, R.T. (Eds.), Hormones, Brain and Behavior, Vol. 1. Academic Press, New York, pp. 339–392. Simon, N.G., McKenna, S.E., Lu, S.-F., Cologer-Clifford, A., 1996. Development and expression of hormonal systems regulating aggression. Ann. N.Y. Acad. Sci. 794, 8–17. Soma, K.K., Sullivan, K., Wingfield, J.C., 1999. Combined aromatase inhibitor and antiandrogen treatment decreases territorial aggression in a wild songbird during the nonbreeding season. Gen. Comp. Endocrinol. 115, 442–453. Soma, K.K., Sullivan, K., Tramontin, A.D., Saldanha, C.J., Schlinger, B.A., Wingfield, J.C., 2000a. Acute and chronic effects of an aromatase inhibitor on territorial aggression in breeding and nonbreeding male song sparrows. J. Comp. Physiol., A 186, 759–769. Soma, K.K., Tramontin, D., Wingfield, J.C., 2000b. Oestrogen regulates male aggression in the non-breeding season. Proc. R. Soc. Lond., B 267, 1089–1096. Sperry, T.S., Thomas, P., 1999. Characterization of two nuclear androgen receptors in Atlantic croaker: comparison of their biochemical properties and binding specificities. Endocrinology 140, 1602–1611. Sperry, T.S., Thompson, C.K., Wingfield, J.C., 2003. Effects of acute treatment with 8-OHDPAT and fluoxetine on aggressive behaviour in male song sparrows (Melospiza melodia morphna). J. Neuroendocrinol. 15, 150–160. Sperry, T.S., Moore, I.T., Meddle, S.L., Benowitz-Fredericks, Z.M., Wingfield, J.C., 2005. Increased sensitivity of the serotonergic system during the breeding season in freeliving American tree sparrows. Behav. Brain Res. 157, 119–126. Thomas, P., Dressing, G., Pang, Y., Berg, H., Tubbs, C., Benninghoff, A., Doughty, K., 2006. Progestin, estrogen and androgen G-protein coupled receptors in fish gonads. Steroids 71, 310–316. Tokarz, R.R., 1987. Effects of the antiandrogens cyproterone acetate and flutamide on male reproductive behavior in a lizard (Anolis sagrei). Horm. Behav. 21, 1–16. Vagell, M.E., McGinnis, M.Y., 1998. The role of gonadal steroid receptor activation in the restoration of sociosexual behavior in adult male rats. Horm. Behav. 33, 163–179. Van Roo, B.L., 2004. Exogenous testosterone inhibits several forms of male parental behavior and stimulates song in a monogamous songbird: the blue-headed vireo (Vireo solitarius). Horm. Behav. 46, 678–683. Wade, J., Buhlman, L., 2000. Lateralization and effects of adult androgen in a sexually dimorphic neuromuscular system controlling song in zebra finches. J. Comp. Neurol. 426, 154–164. Wilson, C.M., McPhaul, M.J., 1996. A and B forms of the androgen receptor are expressed in a variety of human tissues. Mol. Cell. Endocrinol. 120, 51–57. Wingfield, J.C., 1985. Environmental and endocrine control of territorial behavior in

T.S. Sperry et al. / Hormones and Behavior 57 (2010) 86–95 birds. In: Follett, B.K., Ishii, S., Chandola, A. (Eds.), The Endocrine System and the Environment. Springer-Verlag, Berlin, pp. 265–277. Wingfield, J.C., 1993. Control of testicular cycles in the song sparrow, Melospiza melodia melodia: interaction of photoperiod and an endogenous program? Gen. Comp. Endocrinol. 92, 388–401. Wingfield, J.C., 1994. Regulation of territorial behavior in the sedentary song sparrow, Melospiza melodia morphna. Horm. Behav. 28, 1–15. Wingfield, J.C., Farner, D.S., 1978. The endocrinology of a naturally breeding population of the white-crowned sparrow (Zonotrichia leucophrys pugetensis). Physiol. Zool. 51, 188–205. Wingfield, J.C., Hahn, T.P., 1994. Testosterone and territorial behavior in sedentary and migratory sparrows. Anim. Behav. 47, 77–89. Wingfield, J.C., Monk, D., 1992. Control and context of year-round territorial aggression in the non-migratory song sparrow Zonotrichia melodia morphna. Ornis Scand. 23, 298–303.

95

Wingfield, J.C., Hegner, R.E., Dufty, A.M.J., Ball, G.F., 1990. The “challenge hypothesis”: theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am. Nat. 136, 829–846. Wingfield, J.C., Hegner, R.E., Lewis, D., 1991. Circulating levels of luteinizing hormone and steroid hormones in relation to social status in the cooperatively breeding white-browed sparrow weaver, Plocepasser mahali. J. Zool. Lond. 225, 43–58. Wingfield, J.C., Jacobs, J.D., Soma, K., Maney, D.L., Hunt, K., Wisti-Peterson, D., Meddle, S., Ramenofsky, M., Sullivan, K., 1999. Testosterone, aggression, and communication: ecological bases of endocrine phenomena. In: Konishi, M., Hauser, M. (Eds.), The Biology of Communication. MIT Press, Cambridge, MA, pp. 255–283. Wingfield, J.C., Moore, I.T., Goymann, W., Wacker, D.W., Sperry, T.S., 2006. Contexts and ethology of vertebrate aggression: implications for the evolution of hormonebehavior interactions. In: Nelson, R.J. (Ed.), Biology of Aggression. Oxford University Press, New York, pp. 179–210.