- Email: [email protected]
Evidence for androgen independence of male mounting behavior in white-crowned sparrows (Zonotrichia leucophrys gambelii)
HORMONES
AND
BEHAVIOR
17, 414-423 (1983)
Evidence for Androgen Independence of Male Mounting Behavior in White-Crowned Sparrows (Zonotrichia leucophrys gambelii) MICHAEL C. MOORE’ AND RHONDA KRANZ’ Department of zoology, University of Washington, Seattle, Washington, 98195 Previous experiments have shown that expression of mounting behavior in sexually inexperienced, adult male white-crowned sparrows does not require elevated plasma levels of androgen; adult males maintained on nonstimulatory short days mount sexually receptive females. The experiments reported here demonstrate that (1) sexually inexperienced, prepubertal males maintained on nonstimulatory short days show very low mounting rates in response to female sexual displays; (2) these males exhibit high mounting rates when exposed to stimulatory long days but androgen treatment on short days is ineffective in stimulating high mounting rates; and (3) prepubertal castration has no effect on the expression of mounting behavior by photostimulated adult males. Thus, there is no evidence that mounting behavior of adult male white-crowned sparrows depends on androgen.
It is generally accepted that copulatory behavior of male birds is androgen dependent (recent reviews by Arnold 1975; Silver, O’Connell, and Saad, 1979; Adkins-Regan, 1981). However, surprisingly few species, and almost exclusively domestic forms, have been investigated in this regard. Castration has been shown to abolish copulatory behavior (as distinct from courtship behavior) in adult males of only six species of birds: domestic pigeon (Columbia livia; Carpenter, 1933; Erpino, 1969), ring dove (Streptop&a risoria; Barfield, 1971; McDonald and Liley, 1978), mallard (Anas platyrhynchos; Phillips and McKinney, 1962), japanese quail (Coturnix cotumix; Beach and Inman, 1965), domestic chicken (Gallus domesticus; Bar-field, 1969), and zebra finch (Poephila guttata; Arnold, 1975). For a few other species, exogenous androgen is known to stimulate copulatory behavior in immature or sexually inactive males (see above reviews). Recent experiments indicate that male white-crowned sparrows (Zon’ Present address: Department of Zoology, Arizona State University, Tempe, Ariz. 85287. * Present address: Department of Biology, University of Pennsylvania,Philadelphia,Pa.
19104. 414 0018-506X/83 $1.50 Copyright All tights
8 1983 by Academic Press, Inc. of reproduction in any form reserved.
SEXUAL
BEHAVIOR
OF MALE
SPARROWS
415
otrichia leucophrys gambelli) appear to be an exception to the above generalization. White-crowned sparrows are strongly photoperiodic; males maintained on nonstimulatory, short daylengths have very small, nonfunctional testes and basal levels of androgens (Wingfield and Famer, 1980) even when exposed to sexually receptive females (Moore, 1983). These “functionally castrated” males mount and copulate with sexually receptive females as frequently as photostimulated males (Moore, 1983). Although this appears to be the first evidence suggesting androgen independence of mounting behavior in a bird, there are several species in other vertebrate classes for which male copulatory behavior does not appear to depend on androgens (for review see Moore, 1983; Crews, submitted for publication). The most well-documented example is the red-sided garter snake (Thamnophis sirtalis; Garstka, Camazine, and Crews, 1982). Although the above cited experiments have established that expression of mounting behavior in sexually inexperienced adult male white-crowned sparrows does not require elevated plasma levels of androgens, little is known of the effects of androgen on development of male sexual behavior in this species. The following experiments were conducted to test further the androgen independence of mounting behavior of adult males and to examine the effects of androgen treatment on its development in prepubertal males. Although white-crowned sparrows attain adult body size at an age of less than 1 month, initial gonadal development (puberty) does not occur until exposure to stimulatory long days (see Wingfield and Farner, 1980; Wingfield, Smith, and Farner, 1980, for review). For purposes of describing the following experiments, adult males are defined as those that have been photostimulated for at least 25-30 days, a treatment sufficient to induce complete testicular development; prepubertal or immature males are defined as those that have not previously been photostimulated. Previous experiments have shown that prepubertal, firstyear males that have been transferred to stimulatory long days for 25 30 days mount at a high rate (Moore, 1983). The following experiments ( 1) compared mounting rates in untreated and androgen-treated prepubertal males held on short days and (2) determined the effect of prepubertal castration on expression of mounting behavior in photostimulated, adult males. MATERIALS
AND METHODS
Experiment I. Effect of androgen on mounting behavior of prepubertal males. White-crowned sparrows were captured from migratory flocks in
Yakima County, Washington, in September and October 1980. They were sexed by examining the gonads during laparotomy and placed initially in outdoor aviaries in Seattle. They were then transferred indoors on 4 November 1980 to controlled environment chambers and were held two
416
MOOREANDKRANZ
birds of the same sex per cage (41 x 26 x 22 cm) at 22°C and 55% relative humidity on a nonstimulatory photoregimen (8 L: 16 D). Fluorescent lamps provided at least 400 Ix of illumination at the floor of each cage. Food (chick starter mash) and water were always available. First-year males and adult females (too few first-year females were captured) were randomly sorted into bisexual pairs on 16 November 1980 and assigned to either the control (N = 8 pairs) or the experimental (N = 7) group. Each group was placed in a separate environmental chamber and each pair was visually isolated from other pairs with opaque dividers. On Day 0 (16 November 1980) three Silastic capsules (i.d. 1.46 mm, o.d. 1.97 mm, length 14 mm) containing crystalline estradiol were implanted subcutaneously into each female. This number of implants is known to produce maximum physiological levels of estradiol and to make the females sexually receptive (Moore, 1983).Each experimental male similarly received a Silastic capsule (i.d. I .46 mm, o.d. 1.97 mm, length 2.5 mm) containing crystalline testosterone. Pilot studies indicated that these capsules would produce testosterone levels near the maximum of captive photostjmulated males. Each control male received an empty capsule of Identical dimensions. On Days 20-27 the frequencies of female precopulatory displays and male mounts were recorded for 15 min a day for each pair (see Moore, 1983, for description of behaviors). Observations were made through one-way mirrors between 1 and 4 hr after lights came on. Following completion of the observations on Day 27, a blood sample (about 500 ~1) was collected from a wing vein of each male; plasma was separated by centrifugation and stored at -20°C until assayed. Plasma levels of testosterone were measured by radioimmunoassay after extraction of plasma samples with dichloromethane and chromatography on Celite : propylene glycol : ethylene glycol columns. For further details and validation of these assays see Wingfield and Farner (1975) and Moore (1982). Experiment II. Effect of prepubertal castration on mounting behaivior. First-year males were captured in December 1978 and moved indoors to environmental chambers in January 1979 and held on 8 L: 16 D. Four males were bilaterally castrated on 19 April 1979 and six controls were subjected to a sham operation. Following these operations, but prior to being used in this experiment, the males were exposed to a stimulatory daylength (20 L:4 D) from June 1979 to April 1980 and again from November to December 1980. Thus all males had gone through two complete photostimulated cycles of gonadotropin secretion, and intact males had completed two cycles of testicular growth and testosterone secretion, before being used in this experiment (cf. Wingfield and Farner, 1980). It is important to note that these males were never exposed to females after they were brought indoors as first-year birds and therefore
SEXUAL
BEHAVIOR
OF MALE
SPARROWS
417
were sexually inexperienced as adults. Before being used in this experiment, all males were exposed to 6 weeks of 8 L: 16 D to insure that they were photosensitive. Ten adult females were captured in October 1980, transferred indoors on 24 December 1980, and maintained on 8 L: 16 D. All subjects were randomly sorted into bisexual pairs on 31 December 1980. On 9 January 1981 all females received three estradiol capsules as in Experiment I. All birds were transferred to 20 L:4 D on Day 0 (29 January 1981). Sexual behavior of each pair was recorded for 90 mm/day (5 pairs could be observed simultaneously) on Days 15, 16, and 17. Males were bled to death (about 1.2 ml of blood) through the jugular vein on Day 27. Plasma storage and assay were as described in Experiment I. All castrated males were dissected and a visual examination for regenerated testicular fragments was made. Unless otherwise stated all conditions in Experiment II were identical to those of Experiment I. RESULTS Although the testosterone implants given to prepubertal males in Experiment I produced the expected difference in testosterone levels between the control males (d = 0.03 &ml ? 0.01 SIX) and testosterone-treated males (x = 1.09 rig/ml & 0.08 SE), mounting frequencies were extremely low in both groups. Only six mounts by control males (a mean of 0.4 mounts/hr) and seven mounts by testosterone-treated males (a mean of 0.6 mounts/hr) were observed. However, estradiol-treated females caged with testosterone-treated males solicited at a rate (x = 21.9 displays/hr ? 13.4 SE) that was significantly (P < 0.01, Wilcoxon Rank-sum Test) less than that of females caged with control males (x = 56.0 display&r f 8.4 SE). This is apparently because the increased aggressiveness of testosterone-treated males inhibits solicitation displays by females (M. C. Moore and R. Kranz, unpublished data). Nevertheless, because females in both groups solicited at high rates, the very low mounting rates indicate a low responsiveness of these males to female sexual displays. In Experiment II, complete castration was validated for three of the four males because they had testosterone levels below the sensitivity of the assay (co.05 rig/ml) and no visible testicular fragments. The fourth male had measurable plasma levels of testosterone (0.14 rig/ml) and a visible testicular fragment. Therefore, the latter male was excluded from the following analysis. Testosterone levels of intact males were elevated to levels (x = 0.98 rig/ml -+ 0.20 SE) comparable to those in previous experiments (Moore, 1983). Both castrated and control males mounted frequently. The fraction of males observed to mount (four of six controls and two of three castrates) was identical in both treatment groups and mounting frequencies for castrated males (x = 5.9 mounts/hr 2 2.9 SE) were not significantly
418
MOOREANDKRANZ
different from those for control males (x = 4.1 mounts/hr ? 0.9 SE; P > 0.5, Wilcoxon rank-sum Test). Similarly, females caged with castrated males solicited at a rate (J? = 7.6 display&r -+ 2.9 SE) that was statistically indistinguishable from control females (x = 10.0 displays/hr + 2.3 SE; P > 0.4, Wilcoxon rank-sum Test). The much lower solicitation display rates of females in this experiment relative to those in Experiment I may reflect the increased responsiveness of males paired with the former. DISCUSSION
Previous studies have demonstrated that expression of mounting behavior by male white-crowned sparrows is not correlated with plasma levels of testosterone. Males maintained on short day mount as frequently as males maintained on long days, even though testosterone levels of the latter are about lOO-fold greater (Moore, 1983). This suggests that mounting behavior may not be androgen dependent in this species. Classically, two experimental paradigms have been used to demonstrate androgen dependence of male mounting behavior (see references in the introduction). These are (1) castration of sexually active males followed by androgen replacement and (2) administration of androgen to sexually immature or sexually inactive males. The results of the present study indicate that, in both of these classical paradigms, male white-crowned sparrows respond as would be expected if male mounting behavior is androgen independent. In Experiment I, androgen was ineffective in activating mounting behavior in fully grown, prepubertal males. Although this ineffectiveness might be due to immaturity of androgen-responsive tissues, this explanation seems unlikely because photostimulated males of the same age as those in Experiment I mount at high rates (Moore, 1983). In Experiment 11, long-term surgical castration had no apparent effect on mounting rates of adult males. Admittedly, sample sizes in this experiment were small. However, previous studies of other species of birds have shown that long-term castration completely abolishes mounting in every or nearly every individual. Despite the small sample sizes, it is clear from Experiment II that castration does not have the same effect on male white-crowned sparrows. Therefore, at the very least, male white-crowned sparrows differ from other species studied so far in the relative dependence of mounting behavior on gonadal factors. Although results obtained so far indicate that expression of mounting behavior by male white-crowned sparrows does not require elevated plasma levels of androgen, the possibility that androgen plays some role cannot yet be ruled out completely. For example, it is conceivable that very low levels of androgen from some nongonadal source, such as the adrenal gland, are sufficient to activate mounting behavior. It is also possible that androgen has a small effect on frequency of mounting. Such small effects could not be detected without impractically large sample
SEXUAL
BEHAVIOR
OF MALE
SP.4RROWS
419
sizes because of high variance in the behavioral response of captive individuals of this genetically variable, wild species (coefficients of variation for mounting frequencies typically range from 60-120%; present study. Moore, 1983, 1984). However, there have been no consistent trends in several replicates (see below) for higher mounting rates in males maintained on long days compared to males maintained on short days, as would be expected if androgen is affecting mounting rates. Although neither androgen nor photoperiod appears to play a major role in regulating mounting behavior in adult white-crowned sparrows, comparison of all results obtained so far suggests that photoperiod plays an important developmental role. Because of the previously mentioned high variance in behavioral responses of this species in captivity, it is only possible at this time to make a qualitative comparison of results. Nevertheless, this comparison, presented in Table 1. seems clearcut. Mounting frequencies of males fall into two distinct categories: males in one category, which we have designated as “high” mounting frequencies, have frequencies nearly 10 times greater than males with ‘ilow” frequencies. Similarly, in all experiments conducted so far, males have had either basal androgen levels or fully elevated levels about loo-fold greater. From Table 1, it appears that the initial photostimulation in a male’s life initiates high mounting frequencies that are not subsequently affected by changes in photoperiod or plasma levels of androgen. This initiation of high mounting rates by photostimulation could not be mimicked by androgen treatment of prepubertal, i.e., previously nonphotostimuiated. males in Experiment I of the present study nor could it be blocked by castration prior to initial photostimulation in Experiment II. Taken together, these data suggest that initial exposure of prepubertal males to long days, either by direct action on the nervous system or indirectly by stimulating some nongonadal factor, has a permanent, organizational effect on male sexual behavior. Although early exposure to gonadal steroid hormones is known to affect the organization of sexual behavior in several bird species (Adkins, 1978; Konishi and Gurney, 1982), we believe our data are the first to suggest that photoperiod may play a role, independent of its effects on the gonad. in this process (cf. Hutchison, 1974; McDonald, 1982). However, experiments designed specifically to test this interpretation are necessary before it can be fully accepted. Adult male white-crowned sparrows differ from other species of birds in which either short day treatment (Sachs, 1969) or castration (see the introduction) abolishes male mounting behavior. However, there does appear to be variability among species in the dependence of male copulatory behavior on elevated androgen levels (see reviews in Silver et al., 1979; Harding, 1981; Dittami, 1981). For example, male japanese quail appear to require elevated plasma levels of androgen for the expression of cop-
High High High High High High Low Low
Mounting rate’ High High High Low Low Low Low High
Testosterone level* Initial photostimulation Initial photostimulation Third photostimulation Third photostimulation Previous photostimulation Previous photostimulation No previous photostimulation No previous photostimulation
Photostimulation history -..
Source Moore (1983) Moore ( 1984) Present study Present study Moore ( 1983) Moore and Kranz, unpublished Present study Present study
’ High mounting rates range from means of 4.1 to 9.2 mountlhr and low mounting rates from 0.4 to 0.6 mount/hr. * High testosterone levels range from means of 0.98 to 1.62 rig/ml; low testosterone levels are those less than 0.0.5 rig/ml.
Long-day intact (12) Long-day intact (12) Long-day intact (6) Long-day castrate (3) Short-day intact (8) Short-day intact (6) Short-day intact (8) Short-day intact (7), testosterone treated
Condition (N)
TABLE I Comparison of Mounting Rates of Sexually Inexperienced Males in Various Photoperiodic and Endocrine States in Relation to Their History of Previous Photostimulation
N
s $
z
i!
5
SEXUAL
BEHAVIOR
OF MALE SPARROWS
421
ulatory behavior (Beach and Inman, 1965; Sachs, 1969; Adkins-Regan, 1981). Similarly, copulatory behavior of male domestic mallards is correlated with plasma levels of testosterone (Balthazart and Hendrick, 1976). In contrast, mounting behavior of male ring doves appears to require only relatively low levels of androgen. Males mount females during all phases of the reproductive cycle and expression of this behavior is not correlated with plasma levels of androgens (Feder. Storey, Goodwin, Reboulleau, and Silver, 1977; Silver and Barbaire, 1977; Cheng, 1979). Furthermore, males of this weakly photoperiodic species mount receptive females under short-day treatment, although this response is abolished by castration (McDonald and Liley, 1978; McDonald, 1982). The only other species besides the white-crowned sparrow for which there is suggestive evidence of a lack of androgen dependence of male copulatory behavior is the purple-throated Carib hummingbird (Eulampis jugularis). A field study showed that males with regressed testes mount females during the nonbreeding season (Wolf, 1975). However, hormonal studies are necessary to confirm that this actually represents androgen independence, because some species, e.g., western gulls (Larus occidentalis), have high circulating levels of androgen during the nonbreeding season even though testes are regressed (Wingfield, Newman, Hunt, and Farner, 1982). Unfortunately the lack of comparative data on hormonal dependence of behavior in birds with various reproductive tactics makes it difficult to determine (1) if androgen independence is rare and (2) whether it is related to some aspect of the ecology or behavior of individual species, as has been suggested for the white-crowned sparrow (Moore, 1983, 1984). Further studies of other naturally occurring species with different reproductive tactics are necessary before specific hypotheses for the adaptive significance of coupling behavior to hormonal control can be evaluated. ACKNOWLEDGMENTS We thank K. S. Matt for performing the castrations. D. S. Farner for valuable guidance and suggestions concerning these experiments, and D. Crews for comments on earlier drafts of this manuscript. These experiments were supported by NSF Grants PCM 7717690 and PCM 80-18019 to D. S. Farner and an NIH predoctoral training grant to M. C. Moore
REFERENCES Adkins. E. K. (1978). Sex steroids and the differentiation of avian reproductive behavior. Amer. Zoo/. 18, 501-509. Adkins-Regan, E. (1981). Hormone specificity. androgen metabolism, and social behavior. Amer. Zool. 21, 257-271. Arnold, A. P. (1975). The effects of castration and androgen replacement on song, courtship, and aggression in zebra finches (Poephila g~crturu). J. Exp. Zool. 191, 309-326. Bahhazart, J.. and Hendrick, J. (1976).Annual variation in reproductive behavior. testosterone. and plasma FSH levels in the rouen duck (Anus plu~rh.wrhos). Gen. Camp. En&,uinol. 28, 171-183.
422
MOORE AND KRANZ
Barfield, R. J. (1969). Activation of copulatory behavior by androgen implanted into the preoptic area of the male fowl. Harm. Behav. 1, 37-52. Barfield, R. J. (1971). Activation of sexual and aggressive behavior by androgen implanted into the male ring dove brain. Endocrinology 89, 1470-1476. Beach, F. A., and Inman, N. G. (1965). Effects of castration and androgen replacement on mating in male quail. PNAS 54, 1426-1431. Carpenter, C. R. (1933). Psychobiological studies of social behavior in aves. I. The effect of complete and partial gonadectomy on the primary sexual activity in the male pigeon. .I. Comp. Psychol.
16, 25-57.
Cheng, M.-F. (1979). Progress and prospects in ring dove research: A personal view. Advan. Study Behav. 9, 97-129. Crews, D. Gamete production, mating behavior, and sex hormone production uncoupled. Submitted for publication. Dittami, J. P. (1981). Seasonal changes in the behavior and plasma titers of various hormones in barheaded geese, Anser indicus. J. Tierpsychol. 55, 289-324. Erpino, M. J. (1969). Hormonal control of courtship behavior in the pigeon (Columbia livia). Anim. Behav. 17, 401-405. Feder, H. H., Storey, A., Goodwin, D., Reboulleau, C., and Silver, R. (1977). Testosterone and “5a-dihydrotestosterone” levels in peripheral plasma of male and female ring doves (Streptopeka risoriu) during the reproductive cycle. Biol. Reprod. 16, 666-677. Garstka, W. R., Camazine, B., and Crews, D. (1982). Interactions of behavior and physiology during the annual reproductive cycle of the red-sided garter snake (Thamnophis sirtalis parietalis).
Herpetologica
38, 104-123.
Harding, C. F. (1981). Social modulation of circulating hormone levels in the male. Amer. Zool.
21, 223-231.
Hutchison, J. B. (1974). Effect of photoperiod on the decline in behavioural responsiveness to intra-hypothalamic androgen in doves (Streptopelia risoria). J. Endocrinol. 63, 583584. Konishi, M., and Gurney, M. E. (1982). Sexual differentiation of brain and behaviour. Trends Neurosci.
(Per. Ed.) 5, 20-23.
McDonald, P. A. (1982). Influence of pinealectomy and photoperiod on courtship and nestbuilding in male doves. Physiol. Behav. 29, 813-818. McDonald, P. A., and Liley, N. R. (1978). The effects of photoperiod on androgen-induced reproductive behavior in male ring doves, Streptopelin risoria. Horm. Behav. 10, 8596. Moore, M. C. (1982). Hormonal response of free-living male whitecrowned sparrows to experimental manipulation of female sexual behavior. Horm. Behav. 16, 323-329. Moore, M. C. (1983). Effect of female sexual displays on the endocrine physiology and behaviour of male white-crowned sparrows Zonotrichia leucophrys. J. Zool. (London) 199, 137-148. Moore, M. C. (1984). Changes in territorial defense produced by changes in circulating levels of testosterone: A possible hormonal basis for mate-guarding behavior in whitecrowned sparrows. Behaviour, in press. Phillips, R. E., and McKinney, F. (1962). The role of testosterone in the displays of some ducks. Anim. Behav. 10, 244-246. Sachs, B. D. (1969). Photoperiodic control of reproductive behavior and physiology of the male Japanese quail (Coturnix coturnix japonica). Horm. Behnv. 1, 7-24. Silver, R., and Barbaire, C. (1977). Display of courtship and incubation behavior during the reproductive cycle of the male ring dove (Streptopeka risoria). Horm. Behav. 8, 8-21.
Silver, R., O’Connell, M., and Saad, R. (1979). Effect of androgens on the behavior of birds. In C. Beyer (Ed.), Endocrine Control of Sexual Behavior, pp. 223-278. Raven Press, New York.
SEXUAL BEHAVIOR
OF MALE SPARROWS
423
Wingfield, J. C., and Famer, D. S. (1975). The determination of five steroids in avian plasma by radioimmunoassay and competitive protein-binding. Steroids 26, 31l-327. Wingfield, J. C.. and Famer, D. S. (1980). Control of seasonal reproduction in temperatezone birds. Prog. Reprod. Biol. 5, 62-101. Wingfield, J. C., Newman, A. L., Hunt, G. L., and Famer, D. S. (1982). Endocrine aspects of female-female pairing in the western gull (Larus occidentalis wymani). Anim. Behav. 30, 9-22. Wingfield, J. C., Smith, J. P., and Famer, D. S. (1980). Changes in plasma levels of luteinizing hormone, steroid, and thyroid hormones during the postfledging development of white-crowned sparrows. Zonotrichia leucophtys. Gen. Comp. Endocrinol. 41, 372377.
Wolf, L. L. (1975). “Prostitution” 144.
behavior in a tropical hummingbird. Condor 77, 140-
AND
BEHAVIOR
17, 414-423 (1983)
Evidence for Androgen Independence of Male Mounting Behavior in White-Crowned Sparrows (Zonotrichia leucophrys gambelii) MICHAEL C. MOORE’ AND RHONDA KRANZ’ Department of zoology, University of Washington, Seattle, Washington, 98195 Previous experiments have shown that expression of mounting behavior in sexually inexperienced, adult male white-crowned sparrows does not require elevated plasma levels of androgen; adult males maintained on nonstimulatory short days mount sexually receptive females. The experiments reported here demonstrate that (1) sexually inexperienced, prepubertal males maintained on nonstimulatory short days show very low mounting rates in response to female sexual displays; (2) these males exhibit high mounting rates when exposed to stimulatory long days but androgen treatment on short days is ineffective in stimulating high mounting rates; and (3) prepubertal castration has no effect on the expression of mounting behavior by photostimulated adult males. Thus, there is no evidence that mounting behavior of adult male white-crowned sparrows depends on androgen.
It is generally accepted that copulatory behavior of male birds is androgen dependent (recent reviews by Arnold 1975; Silver, O’Connell, and Saad, 1979; Adkins-Regan, 1981). However, surprisingly few species, and almost exclusively domestic forms, have been investigated in this regard. Castration has been shown to abolish copulatory behavior (as distinct from courtship behavior) in adult males of only six species of birds: domestic pigeon (Columbia livia; Carpenter, 1933; Erpino, 1969), ring dove (Streptop&a risoria; Barfield, 1971; McDonald and Liley, 1978), mallard (Anas platyrhynchos; Phillips and McKinney, 1962), japanese quail (Coturnix cotumix; Beach and Inman, 1965), domestic chicken (Gallus domesticus; Bar-field, 1969), and zebra finch (Poephila guttata; Arnold, 1975). For a few other species, exogenous androgen is known to stimulate copulatory behavior in immature or sexually inactive males (see above reviews). Recent experiments indicate that male white-crowned sparrows (Zon’ Present address: Department of Zoology, Arizona State University, Tempe, Ariz. 85287. * Present address: Department of Biology, University of Pennsylvania,Philadelphia,Pa.
19104. 414 0018-506X/83 $1.50 Copyright All tights
8 1983 by Academic Press, Inc. of reproduction in any form reserved.
SEXUAL
BEHAVIOR
OF MALE
SPARROWS
415
otrichia leucophrys gambelli) appear to be an exception to the above generalization. White-crowned sparrows are strongly photoperiodic; males maintained on nonstimulatory, short daylengths have very small, nonfunctional testes and basal levels of androgens (Wingfield and Famer, 1980) even when exposed to sexually receptive females (Moore, 1983). These “functionally castrated” males mount and copulate with sexually receptive females as frequently as photostimulated males (Moore, 1983). Although this appears to be the first evidence suggesting androgen independence of mounting behavior in a bird, there are several species in other vertebrate classes for which male copulatory behavior does not appear to depend on androgens (for review see Moore, 1983; Crews, submitted for publication). The most well-documented example is the red-sided garter snake (Thamnophis sirtalis; Garstka, Camazine, and Crews, 1982). Although the above cited experiments have established that expression of mounting behavior in sexually inexperienced adult male white-crowned sparrows does not require elevated plasma levels of androgens, little is known of the effects of androgen on development of male sexual behavior in this species. The following experiments were conducted to test further the androgen independence of mounting behavior of adult males and to examine the effects of androgen treatment on its development in prepubertal males. Although white-crowned sparrows attain adult body size at an age of less than 1 month, initial gonadal development (puberty) does not occur until exposure to stimulatory long days (see Wingfield and Farner, 1980; Wingfield, Smith, and Farner, 1980, for review). For purposes of describing the following experiments, adult males are defined as those that have been photostimulated for at least 25-30 days, a treatment sufficient to induce complete testicular development; prepubertal or immature males are defined as those that have not previously been photostimulated. Previous experiments have shown that prepubertal, firstyear males that have been transferred to stimulatory long days for 25 30 days mount at a high rate (Moore, 1983). The following experiments ( 1) compared mounting rates in untreated and androgen-treated prepubertal males held on short days and (2) determined the effect of prepubertal castration on expression of mounting behavior in photostimulated, adult males. MATERIALS
AND METHODS
Experiment I. Effect of androgen on mounting behavior of prepubertal males. White-crowned sparrows were captured from migratory flocks in
Yakima County, Washington, in September and October 1980. They were sexed by examining the gonads during laparotomy and placed initially in outdoor aviaries in Seattle. They were then transferred indoors on 4 November 1980 to controlled environment chambers and were held two
416
MOOREANDKRANZ
birds of the same sex per cage (41 x 26 x 22 cm) at 22°C and 55% relative humidity on a nonstimulatory photoregimen (8 L: 16 D). Fluorescent lamps provided at least 400 Ix of illumination at the floor of each cage. Food (chick starter mash) and water were always available. First-year males and adult females (too few first-year females were captured) were randomly sorted into bisexual pairs on 16 November 1980 and assigned to either the control (N = 8 pairs) or the experimental (N = 7) group. Each group was placed in a separate environmental chamber and each pair was visually isolated from other pairs with opaque dividers. On Day 0 (16 November 1980) three Silastic capsules (i.d. 1.46 mm, o.d. 1.97 mm, length 14 mm) containing crystalline estradiol were implanted subcutaneously into each female. This number of implants is known to produce maximum physiological levels of estradiol and to make the females sexually receptive (Moore, 1983).Each experimental male similarly received a Silastic capsule (i.d. I .46 mm, o.d. 1.97 mm, length 2.5 mm) containing crystalline testosterone. Pilot studies indicated that these capsules would produce testosterone levels near the maximum of captive photostjmulated males. Each control male received an empty capsule of Identical dimensions. On Days 20-27 the frequencies of female precopulatory displays and male mounts were recorded for 15 min a day for each pair (see Moore, 1983, for description of behaviors). Observations were made through one-way mirrors between 1 and 4 hr after lights came on. Following completion of the observations on Day 27, a blood sample (about 500 ~1) was collected from a wing vein of each male; plasma was separated by centrifugation and stored at -20°C until assayed. Plasma levels of testosterone were measured by radioimmunoassay after extraction of plasma samples with dichloromethane and chromatography on Celite : propylene glycol : ethylene glycol columns. For further details and validation of these assays see Wingfield and Farner (1975) and Moore (1982). Experiment II. Effect of prepubertal castration on mounting behaivior. First-year males were captured in December 1978 and moved indoors to environmental chambers in January 1979 and held on 8 L: 16 D. Four males were bilaterally castrated on 19 April 1979 and six controls were subjected to a sham operation. Following these operations, but prior to being used in this experiment, the males were exposed to a stimulatory daylength (20 L:4 D) from June 1979 to April 1980 and again from November to December 1980. Thus all males had gone through two complete photostimulated cycles of gonadotropin secretion, and intact males had completed two cycles of testicular growth and testosterone secretion, before being used in this experiment (cf. Wingfield and Farner, 1980). It is important to note that these males were never exposed to females after they were brought indoors as first-year birds and therefore
SEXUAL
BEHAVIOR
OF MALE
SPARROWS
417
were sexually inexperienced as adults. Before being used in this experiment, all males were exposed to 6 weeks of 8 L: 16 D to insure that they were photosensitive. Ten adult females were captured in October 1980, transferred indoors on 24 December 1980, and maintained on 8 L: 16 D. All subjects were randomly sorted into bisexual pairs on 31 December 1980. On 9 January 1981 all females received three estradiol capsules as in Experiment I. All birds were transferred to 20 L:4 D on Day 0 (29 January 1981). Sexual behavior of each pair was recorded for 90 mm/day (5 pairs could be observed simultaneously) on Days 15, 16, and 17. Males were bled to death (about 1.2 ml of blood) through the jugular vein on Day 27. Plasma storage and assay were as described in Experiment I. All castrated males were dissected and a visual examination for regenerated testicular fragments was made. Unless otherwise stated all conditions in Experiment II were identical to those of Experiment I. RESULTS Although the testosterone implants given to prepubertal males in Experiment I produced the expected difference in testosterone levels between the control males (d = 0.03 &ml ? 0.01 SIX) and testosterone-treated males (x = 1.09 rig/ml & 0.08 SE), mounting frequencies were extremely low in both groups. Only six mounts by control males (a mean of 0.4 mounts/hr) and seven mounts by testosterone-treated males (a mean of 0.6 mounts/hr) were observed. However, estradiol-treated females caged with testosterone-treated males solicited at a rate (x = 21.9 displays/hr ? 13.4 SE) that was significantly (P < 0.01, Wilcoxon Rank-sum Test) less than that of females caged with control males (x = 56.0 display&r f 8.4 SE). This is apparently because the increased aggressiveness of testosterone-treated males inhibits solicitation displays by females (M. C. Moore and R. Kranz, unpublished data). Nevertheless, because females in both groups solicited at high rates, the very low mounting rates indicate a low responsiveness of these males to female sexual displays. In Experiment II, complete castration was validated for three of the four males because they had testosterone levels below the sensitivity of the assay (co.05 rig/ml) and no visible testicular fragments. The fourth male had measurable plasma levels of testosterone (0.14 rig/ml) and a visible testicular fragment. Therefore, the latter male was excluded from the following analysis. Testosterone levels of intact males were elevated to levels (x = 0.98 rig/ml -+ 0.20 SE) comparable to those in previous experiments (Moore, 1983). Both castrated and control males mounted frequently. The fraction of males observed to mount (four of six controls and two of three castrates) was identical in both treatment groups and mounting frequencies for castrated males (x = 5.9 mounts/hr 2 2.9 SE) were not significantly
418
MOOREANDKRANZ
different from those for control males (x = 4.1 mounts/hr ? 0.9 SE; P > 0.5, Wilcoxon rank-sum Test). Similarly, females caged with castrated males solicited at a rate (J? = 7.6 display&r -+ 2.9 SE) that was statistically indistinguishable from control females (x = 10.0 displays/hr + 2.3 SE; P > 0.4, Wilcoxon rank-sum Test). The much lower solicitation display rates of females in this experiment relative to those in Experiment I may reflect the increased responsiveness of males paired with the former. DISCUSSION
Previous studies have demonstrated that expression of mounting behavior by male white-crowned sparrows is not correlated with plasma levels of testosterone. Males maintained on short day mount as frequently as males maintained on long days, even though testosterone levels of the latter are about lOO-fold greater (Moore, 1983). This suggests that mounting behavior may not be androgen dependent in this species. Classically, two experimental paradigms have been used to demonstrate androgen dependence of male mounting behavior (see references in the introduction). These are (1) castration of sexually active males followed by androgen replacement and (2) administration of androgen to sexually immature or sexually inactive males. The results of the present study indicate that, in both of these classical paradigms, male white-crowned sparrows respond as would be expected if male mounting behavior is androgen independent. In Experiment I, androgen was ineffective in activating mounting behavior in fully grown, prepubertal males. Although this ineffectiveness might be due to immaturity of androgen-responsive tissues, this explanation seems unlikely because photostimulated males of the same age as those in Experiment I mount at high rates (Moore, 1983). In Experiment 11, long-term surgical castration had no apparent effect on mounting rates of adult males. Admittedly, sample sizes in this experiment were small. However, previous studies of other species of birds have shown that long-term castration completely abolishes mounting in every or nearly every individual. Despite the small sample sizes, it is clear from Experiment II that castration does not have the same effect on male white-crowned sparrows. Therefore, at the very least, male white-crowned sparrows differ from other species studied so far in the relative dependence of mounting behavior on gonadal factors. Although results obtained so far indicate that expression of mounting behavior by male white-crowned sparrows does not require elevated plasma levels of androgen, the possibility that androgen plays some role cannot yet be ruled out completely. For example, it is conceivable that very low levels of androgen from some nongonadal source, such as the adrenal gland, are sufficient to activate mounting behavior. It is also possible that androgen has a small effect on frequency of mounting. Such small effects could not be detected without impractically large sample
SEXUAL
BEHAVIOR
OF MALE
SP.4RROWS
419
sizes because of high variance in the behavioral response of captive individuals of this genetically variable, wild species (coefficients of variation for mounting frequencies typically range from 60-120%; present study. Moore, 1983, 1984). However, there have been no consistent trends in several replicates (see below) for higher mounting rates in males maintained on long days compared to males maintained on short days, as would be expected if androgen is affecting mounting rates. Although neither androgen nor photoperiod appears to play a major role in regulating mounting behavior in adult white-crowned sparrows, comparison of all results obtained so far suggests that photoperiod plays an important developmental role. Because of the previously mentioned high variance in behavioral responses of this species in captivity, it is only possible at this time to make a qualitative comparison of results. Nevertheless, this comparison, presented in Table 1. seems clearcut. Mounting frequencies of males fall into two distinct categories: males in one category, which we have designated as “high” mounting frequencies, have frequencies nearly 10 times greater than males with ‘ilow” frequencies. Similarly, in all experiments conducted so far, males have had either basal androgen levels or fully elevated levels about loo-fold greater. From Table 1, it appears that the initial photostimulation in a male’s life initiates high mounting frequencies that are not subsequently affected by changes in photoperiod or plasma levels of androgen. This initiation of high mounting rates by photostimulation could not be mimicked by androgen treatment of prepubertal, i.e., previously nonphotostimuiated. males in Experiment I of the present study nor could it be blocked by castration prior to initial photostimulation in Experiment II. Taken together, these data suggest that initial exposure of prepubertal males to long days, either by direct action on the nervous system or indirectly by stimulating some nongonadal factor, has a permanent, organizational effect on male sexual behavior. Although early exposure to gonadal steroid hormones is known to affect the organization of sexual behavior in several bird species (Adkins, 1978; Konishi and Gurney, 1982), we believe our data are the first to suggest that photoperiod may play a role, independent of its effects on the gonad. in this process (cf. Hutchison, 1974; McDonald, 1982). However, experiments designed specifically to test this interpretation are necessary before it can be fully accepted. Adult male white-crowned sparrows differ from other species of birds in which either short day treatment (Sachs, 1969) or castration (see the introduction) abolishes male mounting behavior. However, there does appear to be variability among species in the dependence of male copulatory behavior on elevated androgen levels (see reviews in Silver et al., 1979; Harding, 1981; Dittami, 1981). For example, male japanese quail appear to require elevated plasma levels of androgen for the expression of cop-
High High High High High High Low Low
Mounting rate’ High High High Low Low Low Low High
Testosterone level* Initial photostimulation Initial photostimulation Third photostimulation Third photostimulation Previous photostimulation Previous photostimulation No previous photostimulation No previous photostimulation
Photostimulation history -..
Source Moore (1983) Moore ( 1984) Present study Present study Moore ( 1983) Moore and Kranz, unpublished Present study Present study
’ High mounting rates range from means of 4.1 to 9.2 mountlhr and low mounting rates from 0.4 to 0.6 mount/hr. * High testosterone levels range from means of 0.98 to 1.62 rig/ml; low testosterone levels are those less than 0.0.5 rig/ml.
Long-day intact (12) Long-day intact (12) Long-day intact (6) Long-day castrate (3) Short-day intact (8) Short-day intact (6) Short-day intact (8) Short-day intact (7), testosterone treated
Condition (N)
TABLE I Comparison of Mounting Rates of Sexually Inexperienced Males in Various Photoperiodic and Endocrine States in Relation to Their History of Previous Photostimulation
N
s $
z
i!
5
SEXUAL
BEHAVIOR
OF MALE SPARROWS
421
ulatory behavior (Beach and Inman, 1965; Sachs, 1969; Adkins-Regan, 1981). Similarly, copulatory behavior of male domestic mallards is correlated with plasma levels of testosterone (Balthazart and Hendrick, 1976). In contrast, mounting behavior of male ring doves appears to require only relatively low levels of androgen. Males mount females during all phases of the reproductive cycle and expression of this behavior is not correlated with plasma levels of androgens (Feder. Storey, Goodwin, Reboulleau, and Silver, 1977; Silver and Barbaire, 1977; Cheng, 1979). Furthermore, males of this weakly photoperiodic species mount receptive females under short-day treatment, although this response is abolished by castration (McDonald and Liley, 1978; McDonald, 1982). The only other species besides the white-crowned sparrow for which there is suggestive evidence of a lack of androgen dependence of male copulatory behavior is the purple-throated Carib hummingbird (Eulampis jugularis). A field study showed that males with regressed testes mount females during the nonbreeding season (Wolf, 1975). However, hormonal studies are necessary to confirm that this actually represents androgen independence, because some species, e.g., western gulls (Larus occidentalis), have high circulating levels of androgen during the nonbreeding season even though testes are regressed (Wingfield, Newman, Hunt, and Farner, 1982). Unfortunately the lack of comparative data on hormonal dependence of behavior in birds with various reproductive tactics makes it difficult to determine (1) if androgen independence is rare and (2) whether it is related to some aspect of the ecology or behavior of individual species, as has been suggested for the white-crowned sparrow (Moore, 1983, 1984). Further studies of other naturally occurring species with different reproductive tactics are necessary before specific hypotheses for the adaptive significance of coupling behavior to hormonal control can be evaluated. ACKNOWLEDGMENTS We thank K. S. Matt for performing the castrations. D. S. Farner for valuable guidance and suggestions concerning these experiments, and D. Crews for comments on earlier drafts of this manuscript. These experiments were supported by NSF Grants PCM 7717690 and PCM 80-18019 to D. S. Farner and an NIH predoctoral training grant to M. C. Moore
REFERENCES Adkins. E. K. (1978). Sex steroids and the differentiation of avian reproductive behavior. Amer. Zoo/. 18, 501-509. Adkins-Regan, E. (1981). Hormone specificity. androgen metabolism, and social behavior. Amer. Zool. 21, 257-271. Arnold, A. P. (1975). The effects of castration and androgen replacement on song, courtship, and aggression in zebra finches (Poephila g~crturu). J. Exp. Zool. 191, 309-326. Bahhazart, J.. and Hendrick, J. (1976).Annual variation in reproductive behavior. testosterone. and plasma FSH levels in the rouen duck (Anus plu~rh.wrhos). Gen. Camp. En&,uinol. 28, 171-183.
422
MOORE AND KRANZ
Barfield, R. J. (1969). Activation of copulatory behavior by androgen implanted into the preoptic area of the male fowl. Harm. Behav. 1, 37-52. Barfield, R. J. (1971). Activation of sexual and aggressive behavior by androgen implanted into the male ring dove brain. Endocrinology 89, 1470-1476. Beach, F. A., and Inman, N. G. (1965). Effects of castration and androgen replacement on mating in male quail. PNAS 54, 1426-1431. Carpenter, C. R. (1933). Psychobiological studies of social behavior in aves. I. The effect of complete and partial gonadectomy on the primary sexual activity in the male pigeon. .I. Comp. Psychol.
16, 25-57.
Cheng, M.-F. (1979). Progress and prospects in ring dove research: A personal view. Advan. Study Behav. 9, 97-129. Crews, D. Gamete production, mating behavior, and sex hormone production uncoupled. Submitted for publication. Dittami, J. P. (1981). Seasonal changes in the behavior and plasma titers of various hormones in barheaded geese, Anser indicus. J. Tierpsychol. 55, 289-324. Erpino, M. J. (1969). Hormonal control of courtship behavior in the pigeon (Columbia livia). Anim. Behav. 17, 401-405. Feder, H. H., Storey, A., Goodwin, D., Reboulleau, C., and Silver, R. (1977). Testosterone and “5a-dihydrotestosterone” levels in peripheral plasma of male and female ring doves (Streptopeka risoriu) during the reproductive cycle. Biol. Reprod. 16, 666-677. Garstka, W. R., Camazine, B., and Crews, D. (1982). Interactions of behavior and physiology during the annual reproductive cycle of the red-sided garter snake (Thamnophis sirtalis parietalis).
Herpetologica
38, 104-123.
Harding, C. F. (1981). Social modulation of circulating hormone levels in the male. Amer. Zool.
21, 223-231.
Hutchison, J. B. (1974). Effect of photoperiod on the decline in behavioural responsiveness to intra-hypothalamic androgen in doves (Streptopelia risoria). J. Endocrinol. 63, 583584. Konishi, M., and Gurney, M. E. (1982). Sexual differentiation of brain and behaviour. Trends Neurosci.
(Per. Ed.) 5, 20-23.
McDonald, P. A. (1982). Influence of pinealectomy and photoperiod on courtship and nestbuilding in male doves. Physiol. Behav. 29, 813-818. McDonald, P. A., and Liley, N. R. (1978). The effects of photoperiod on androgen-induced reproductive behavior in male ring doves, Streptopelin risoria. Horm. Behav. 10, 8596. Moore, M. C. (1982). Hormonal response of free-living male whitecrowned sparrows to experimental manipulation of female sexual behavior. Horm. Behav. 16, 323-329. Moore, M. C. (1983). Effect of female sexual displays on the endocrine physiology and behaviour of male white-crowned sparrows Zonotrichia leucophrys. J. Zool. (London) 199, 137-148. Moore, M. C. (1984). Changes in territorial defense produced by changes in circulating levels of testosterone: A possible hormonal basis for mate-guarding behavior in whitecrowned sparrows. Behaviour, in press. Phillips, R. E., and McKinney, F. (1962). The role of testosterone in the displays of some ducks. Anim. Behav. 10, 244-246. Sachs, B. D. (1969). Photoperiodic control of reproductive behavior and physiology of the male Japanese quail (Coturnix coturnix japonica). Horm. Behnv. 1, 7-24. Silver, R., and Barbaire, C. (1977). Display of courtship and incubation behavior during the reproductive cycle of the male ring dove (Streptopeka risoria). Horm. Behav. 8, 8-21.
Silver, R., O’Connell, M., and Saad, R. (1979). Effect of androgens on the behavior of birds. In C. Beyer (Ed.), Endocrine Control of Sexual Behavior, pp. 223-278. Raven Press, New York.
SEXUAL BEHAVIOR
OF MALE SPARROWS
423
Wingfield, J. C., and Famer, D. S. (1975). The determination of five steroids in avian plasma by radioimmunoassay and competitive protein-binding. Steroids 26, 31l-327. Wingfield, J. C.. and Famer, D. S. (1980). Control of seasonal reproduction in temperatezone birds. Prog. Reprod. Biol. 5, 62-101. Wingfield, J. C., Newman, A. L., Hunt, G. L., and Famer, D. S. (1982). Endocrine aspects of female-female pairing in the western gull (Larus occidentalis wymani). Anim. Behav. 30, 9-22. Wingfield, J. C., Smith, J. P., and Famer, D. S. (1980). Changes in plasma levels of luteinizing hormone, steroid, and thyroid hormones during the postfledging development of white-crowned sparrows. Zonotrichia leucophtys. Gen. Comp. Endocrinol. 41, 372377.
Wolf, L. L. (1975). “Prostitution” 144.
behavior in a tropical hummingbird. Condor 77, 140-