Androgens and dominance: Sex-specific patterns in a highly social fish (Neolamprologus pulcher)

https://doi.org/10.1016/j.ygcen.2008.12.018Get rights and content

Abstract

In most vertebrates, aggression and dominance are tightly linked to circulating testosterone. Fish, however, have two androgens (testosterone, T and 11-ketotestosterone, 11KT) that influence aggression and dominance. To date, few studies have compared the relationship between androgen levels and the outcome of aggressive contests in both females and males of the same species. To investigate sex differences in androgens we staged size-matched, limited-resource (territory) contests with 14 female–female and 10 male–male pairs of the highly social cichlid Neolamprologus pulcher. We then examined androgen levels in recently established dominants, who won the contest and subsequently acquired a territory (for 3 h), and subordinates, who lost and did not acquire a territory. Newly dominant females had higher plasma T but similar 11KT levels to newly subordinate females. In contrast, newly dominant males had higher 11KT but similar T levels to subordinate males. The ratio of 11KT to T, which demonstrates physiological importance of T conversion to 11KT, was positively correlated with submissive behavior in female winners, and correlated weakly with aggressive behavior in male winners (p = 0.05). These findings provide support for the hypothesis that different androgens play equivalent roles in female versus male dominance establishment, and suggest that relative levels of 11KT and T are implicated in female dominance behavior and perhaps behavior of both sexes.

Introduction

Androgen hormones have been linked to aggressive behavior across a broad spectrum of vertebrates (reviewed by Nelson (2000)). Manipulation experiments have demonstrated that androgen removal decreases aggression while androgen treatment rescues or increases aggression (Edwards, 1970, Arnold, 1975, Balthazart, 1983, Moore, 1988, Kindler et al., 1991, Francis et al., 1992, Hume and Wynne-Edwards, 2005). These observations, paired with repeated demonstration that individuals with elevated androgen levels exhibit high levels of aggression (Lincoln et al., 1972, Scott et al., 1980, Johnsen, 1998) have led scientists to accept circulating androgens as potent mediators of male aggressive behavior. More recent findings have shown that female aggression also varies with circulating androgen levels (reviewed in Bouissou, 1983, Staub and De Beer, 1997). Both male and female aggression and androgen levels have been strongly associated with dominance over conspecifics (Bouissou, 1983, Schoech et al., 1991, Desjardins et al., 2008b). To our knowledge, however, no study has measured androgen levels corresponding with dominance establishment in both males and females, and sex-specific roles of androgens in dominance establishment remain poorly understood.

Neolamprologus pulcher, a group-living African cichlid, is an ideal model to test predictions on aggression and dominance establishment in both males and females. Highly territorial, this fish lives in groups composed of a dominant breeding pair and up to 20 subordinate helpers (Balshine et al., 2001, Heg et al., 2005). All individuals constantly reinforce their position within the group dominance hierarchy through a variety of aggressive, submissive and affiliative behaviors. As in all group-living species, predation events and natural deaths create vacancies in the dominance hierarchy. When these vacancies occur in top breeding positions, conspecifics from within or outside the group compete over the newly vacant position (Balshine-Earn et al., 1998, Stiver et al., 2006, Fitzpatrick et al., 2008). The victor of these competitions fills the dominant breeding position. Maintaining a position of dominance requires the strength and stamina for territory defense, subordinate policing and reproduction. These behaviors have been associated with circulating androgen levels in fish (Neat et al., 1998, Parikh et al., 2006), and androgen levels have been implicated in aggressive behavior during male dominance establishment across vertebrate taxa (Bouissou, 1983, Wingfield, 1985, Moore, 1988, Oliveira, 2004). Importantly, a previous experiment with N. pulcher found that winners and losers had equally elevated androgen levels immediately following a contest, although no changes in estradiol were seen in either sex (Desjardins et al., 2006). The androgen increases observed in both contestants may reflect short-term aggression-induced fluctuations rather than long-term patterns associated with status change, as established dominants in the field have higher plasma androgen levels than subordinates (Desjardins et al., 2008b).

The first aim of this experiment was to determine whether individuals successfully acquiring a territory for 3 h after completion of competition would exhibit higher levels of circulating androgen hormones than individuals who were unsuccessful in territory acquisition. The second aim was to explore whether females and males differ in androgen response to territory acquisition and establishment of dominance. To investigate endocrine and behavioral correlates of territory acquisition, we staged competitions for a shelter-containing territory between size-matched individuals of the cichlid fish, N. pulcher. As this fish species is highly philopatric and completely reliant on shelter in its territory for survival and reproduction (Balshine et al., 2001), we staged contests for shelter and then compared plasma androgens, relative gonad investment, and measures of aggression between competition winners and losers. Plasma T and 11-KT were assayed, as these are considered the most active androgens in females and males, respectively (Borg, 1994). Specifically, we predicted to find higher circulating androgen levels in both female and male individuals that successfully gained control of a territory (winners) compared to individuals that did not (losers).

Section snippets

Study animals and holding conditions

Territory competitions were performed between 24 September and 26 November 2004 using N. pulcher held at McMaster University in Hamilton, Ont., Canada. Fish used in this study were derived from a wild-caught stock of fish originating from the southern shore of Lake Tanganyika, received in 2001 and 2002. To mimic natural lake conditions water temperature was kept between 25 and 28 °C and a 13:11 light:dark cycle was maintained. Fish were fed flake cichlid food daily, ad libitum. Tanks contained

Do males and females have similar androgen responses to dominance establishment?

Female individuals that won territory competitions had higher levels of circulating plasma T than those that lost (paired t-test: t7 = −2.46, p < 0.05; Fig. 1A), but there were no male winner–loser differences in plasma T levels (t8 = −1.13, p = 0.29; Fig. 1A).

Females winners and losers did not differ in plasma 11KT levels (t8 = −1.65, p = 0.14; Fig. 1B). Male winners, however, did have higher plasma 11KT than male losers (t6 = −2.90, p < 0.05; Fig. 1B).

The plasma 11KT/T ratio, which indicates the

Discussion

Androgen secretion patterns observed in female and male N. pulcher were in line with previous studies in this and other fish species; circulating T is higher in female fish while 11KT or the 11KT/T ratio is higher in males (Oliveira, 2004, Desjardins et al., 2008b; reviewed in Borg, 1994). Importantly, this is the first study to show that females and males of the same species have distinct hormone profiles associated with recent change in social status. Dominant females had higher T (but not

Acknowledgments

The authors wish to thank Susan Marsh-Rollo, Viktoria Mileva and John Fitzpatrick for their assistance with this work and advice with respect to the statistical analyses. We also thank anonymous referees for comments on the paper. This work was supported by a NSERC Discovery Grant awarded to S.B. and an NSERC doctoral PGS award to J.K.D.

References (39)

Cited by (76)

  • Antioxidant capacity differs across social ranks and with ascension in males of a group-living fish

    2022, Comparative Biochemistry and Physiology -Part A : Molecular and Integrative Physiology
View all citing articles on Scopus
View full text