Testosterone differentially modulates the display of agonistic behavior and dominance over opponents before and after adolescence in male Syrian hamsters

Abstract

The current study investigated the influence of testosterone on agonistic behavior and dominance over an opponent before and after adolescence in male Syrian hamsters (Mesocricetus auratus), and tested the hypothesis that shifts in behavioral responsiveness to testosterone occur across adolescent development. We predicted that testosterone-dependent modulation of attacks decreases following puberty, and that flank marking behavior in response to testosterone increases following puberty. Prepubertal (14 days of age) and adult subjects (52-62 days of age) were gonadectomized and immediately implanted with testosterone propionate (TP) or vehicle pellets. Fourteen days later, agonistic behaviors were assessed in a neutral arena with age-matched testosterone-treated opponents. TP treatment increased attacks and dominance over an opponent in prepubertal but not adult males, supporting the hypothesis that testosterone-dependent modulation of aggression decreases following puberty. TP increased flank marking behavior in adults, but failed to increase flank marking in prepubertal subjects, supporting the hypothesized increase in testosterone-dependent modulation of flank marking after puberty. Thus, we provide here evidence that changes in agonistic responses to steroid hormones occur across puberty and adolescence in male rodents, much like the well-established shifts in neuroendocrine and reproductive behavioral responses to steroid hormones that occur pre- to post-pubertally. These findings may have implications for early pubertal timing and increased risk for externalizing symptoms and aggressive behavior in humans.

Keywords: adolescence; aggressive; agonistic; dominance; puberty; social behavior; submissive; testosterone.

FIGURE 1

Experimental design. The pubertal rise in gonadal secretions of testosterone begins at approximately 30 days of age and reaches adult levels 3 weeks later around 50 days of age. The timing of gonadectomy and behavioral testing procedures were optimized to occur before or after pubertal development. Subjects were gonadectomized either before puberty at 14 days of age, or after puberty between 52 and 62 days. In the same procedure, subjects were implanted with testosterone propionate (TP) or vehicle pellets, resulting in 4 groups: Prepub + 0, Prepub + TP, Adult + 0, and Adult + TP. Behavioral testing occurred 2 weeks later at 28 days of age (Prepub) or 66–76 days of age (Adult). All subjects were singly housed for 1 week prior to scent-induced flank marking tests and social interaction tests. In the scent-induced flank marking test, subjects were placed into a testing aquarium filled with the soiled bedding of gonad-intact adult males and the number of flank marks were quantified. In the social interaction test, subjects were age- and weight-matched with T-treated male opponents, and the number of aggressive and submissive behavioral displays were quantified. Figure created using Biorender.com.

FIGURE 2

Mean body weight and flank gland diameter in castrated prepubertal and adult subjects. Measurements were taken following behavioral testing. (A) Testosterone treatment did not significantly affect body weight in prepubertal or adult subjects. (B) Adults exhibited greater flank gland diameters overall, and testosterone administration significantly increased flank gland diameter in both prepubertal and adult subjects. Circles represent individual values, and the bars represent the group mean. **p < 0.01, ***p < 0.001.

FIGURE 3

Median number [ ± CI 95%] of flank marks displayed by prepubertal and adult subjects during the scent test (A) and social interaction test (B). In both tests, testosterone propionate treatment increased flank marking behavior only in adult subjects. **p < 0.01, ***p < 0.001.

FIGURE 4

The effects of age and hormone on the duration of close contact between subjects and opponents during the social interaction test. (A) Prepubertal subjects spent significantly more time in close contact with opponents than did adults. A significant interaction between Age and Hormone revealed that testosterone propionate (TP) treatment significantly decreased close contact time between adult subjects and opponents, but did not affect prepubertal subject’s time in close contact with opponents. (B) Age and Hormone interacted to affect close contact average bout durations. TP-treatment significantly increased close contact bout durations in prepubertal males and decreased close contact bout durations in adult males. Circles represent individual values, and the bars represent the group mean. *p < 0.05, ***p < 0.001.

FIGURE 5

Median number [ ± CI 95%] of aggressive and submissive behaviors displayed by vehicle- and testosterone propionate (TP)-treated prepubertal and adult subjects during a social interaction with an age-matched opponent. (A) Vehicle-treated prepubertal subjects displayed the highest number of offensive postures, significantly higher than TP-treated adults. (B) TP significantly increased paws-on displays in adult but not prepubertal subjects. (C) TP significantly increased attacks in prepubertal but not adult subjects. Prepub + TP males attacked opponents more often than both Adult + 0 and Adult + TP groups. (D) TP did not influence defensive postures of prepubertal or adult subjects, however, prepubertal subjects displayed higher levels of defensive postures than adults. (E) Tail-up displays were significantly decreased by TP treatment in prepubertal but not in adult subjects. Prepubertal subjects displayed higher levels of tail-up displays than adults. (F) The subject’s aggression score reflects the difference between their average aggressive and submissive displays. Aggression scores were influenced by Age, Hormone, and an interaction between Age and Hormone. TP significantly increased aggression scores in prepubertal males, but not in adult males. (G) The win index score reflects the difference between aggression scores of subjects and their opponents. TP treatment significantly increased dominance scores overall, and particularly in prepubertal males. TP also increased the proportion of prepubertal subjects that won dominance contests. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 6

Opponent attacks and tail up displays during interactions with vehicle- and TP-treated subjects. (A) Adult opponents attacked vehicle-treated subjects more than TP-treated subjects. A significant difference in attacks was also observed between vehicle-treated adult subjects and their opponents, but not between TP-treated adult subjects and opponents. Prepubertal opponent attacks did not differ depending on subject hormone status, nor did attack frequencies differ within subject and opponent dyads. (B) Prepubertal opponents displayed more tail-up postures during interactions with TP-treated subjects than with vehicle-treated subjects. In prepubertal dyads, vehicle-treated subjects displayed significantly more tail-up postures than did their opponents, whereas tail-up displays did not differ between TP-treated prepubertal subjects and their opponents. In adults, opponent tail-up postures did not change depending on subject hormone status, nor were differences observed within subject and opponent dyads. *p < 0.05, **p < 0.01. # denotes a significant difference within subject and opponent dyads.