Androgenic anabolic steroid exposure during adolescence: ramifications for brain development and behavior.

This article is part of a Special Issue "Puberty and Adolescence". Puberty is a critical period for brain maturation that is highly dependent on gonadal sex hormones. Modifications in the gonadal steroid environment, via the use of anabolic androgenic steroids (AAS), have been shown to affect brain development and behavior. Studies in both humans and animal models indicate that AAS exposure during adolescence alters normal brain remodeling, including structural changes and neurotransmitter function. The most commonly reported behavioral effect is an increase in aggression. Evidence has been presented to identify factors that influence the effect of AAS on the expression of aggression. The chemical composition of the AAS plays a major role in determining whether aggression is displayed, with testosterone being the most effective. The hormonal context, the environmental context, physical provocation and the perceived threat during the social encounter have all been found to influence the expression of aggression and sexual behavior. All of these factors point toward an altered behavioral state that includes an increased readiness to respond to a social encounter with heightened vigilance and enhanced motivation. This AAS-induced state may be defined as emboldenment. The evidence suggests that the use of AAS during this critical period of development may increase the risk for maladaptive behaviors along with neurological disorders.

Androgen receptors, sex behavior, and aggression.

Androgens are intricately involved in reproductive and aggressive behaviors, but the role of the androgen receptor in mediating these behaviors is less defined. Further, activity of the hypothalamic-pituitary-gonadal axis and hypothalamic-pituitary-adrenal axis can influence each other at the level of the androgen receptor. Knowledge of the mechanisms for androgens' effects on behaviors through the androgen receptor will guide future studies in elucidating male reproductive and aggressive behavior repertoires.

Changes in circadian rhythms during puberty in Rattus norvegicus: developmental time course and gonadal dependency.

During puberty, humans develop a later chronotype, exhibiting a phase-delayed daily rest/activity rhythm. The purpose of this study was to determine: 1) whether similar changes in chronotype occur during puberty in a laboratory rodent species, 2) whether these changes are due to pubertal hormones affecting the circadian timekeeping system. We tracked the phasing and distribution of wheel-running activity rhythms during post-weaning development in rats that were gonadectomized before puberty or left intact. We found that intact peripubertal rats had activity rhythms that were phase-delayed relative to adults. Young rats also exhibited a bimodal nocturnal activity distribution. As puberty progressed, bimodality diminished and late-night activity phase-advanced until it consolidated with early-night activity. By late puberty, intact rats showed a strong, unimodal rhythm that peaked at the beginning of the night. These pubertal changes in circadian phase were more pronounced in males than females. Increases in gonadal hormones during puberty partially accounted for these changes, as rats that were gonadectomized before puberty demonstrated smaller phase changes than intact rats and maintained ultradian rhythms into adulthood. We investigated the role of photic entrainment by comparing circadian development under constant and entrained conditions. We found that the period (τ) of free-running rhythms developed sex differences during puberty. These changes in τ did not account for pubertal changes in entrained circadian phase, as the consolidation of activity at the beginning of the subjective night persisted under constant conditions in both sexes. We conclude that the circadian system continues to develop in a hormone-sensitive manner during puberty.

Effects of anabolic androgenic steroids and social subjugation on behavior and neurochemistry in male rats.

Early abuse and anabolic androgenic steroids (AAS) both increase aggression. We assessed the behavioral and neurochemical consequences of AAS, alone or in combination with social subjugation (SS), an animal model of child abuse. On P26, gonadally intact male rats began SS consisting of daily pairings with an adult male for 2 weeks followed by daily injections of the AAS, testosterone on P40. As adults, males were tested for sexual and aggressive behaviors towards females in various hormonal conditions and inter-male aggression in a neutral setting using home or opponent bedding. Neurotransmitter levels were assessed using HPLC. Results showed that AAS males displayed significantly more mounts toward sexually receptive, vaginally obstructed females (OBS) and displayed significantly more threats towards ovariectomized females. SS males mounted OBS females significantly less and were not aggressive toward females. The role of olfactory cues in a neutral setting did not affect aggression regardless of treatment. AAS significantly increased brainstem DOPAC and NE. SS decreased 5HIAA, DA, DOPAC, and NE in brainstem. 5HIAA was significantly increased in the prefrontal cortex of all experimental groups. We conclude that AAS and SS differentially affect behavior towards females as well as neurotransmitter levels.

Impact of anabolic androgenic steroids on adolescent males.

Anabolic androgenic steroid (AAS) use increased dramatically among adolescent males. This review focuses on studies using animal models of AAS exposure during adolescence which is a hormonally sensitive developmental period. AAS exposure during this critical period has wide-ranging consequences, including increased dendritic spine density, altered brain serotonin levels and escalated aggression in response to physical provocation. Human data suggest that AAS induces indiscriminate and unprovoked aggression often described as "'roid rage". However, animal studies indicate that the behavioral impact of AAS is modulated by experiential and social contingencies, a perceived provocation, and the chemical composition of the AAS. The AAS, testosterone increases aggression in juvenile and adult male rats when physically provoked. In contrast, stanzolol, inhibits aggression in both juvenile and adult male rats, even when physically provoked. Nandrolone has minimal effects on aggression, unless preceded by attack training. Exposure to AAS during adolescence may have a host of unintended bio-behavioral consequences. Yet, the perception of harmlessness surrounds AAS use. The perception of harmlessness is promoted by the availability of AAS especially through internet pharmacies. The perception of acceptability is reflected in current cultural ethics that no longer condemn cheating to obtain personal achievement or success. A prevailing conviction is that although AAS are illegal they are not really bad. Reduction of the availability of AAS to adolescents requires ardent legislative and legal intervention. The problem of acceptability can be addressed by educating adolescents about the short-term and long-term effects of AAS on brain and behavior, to increase awareness of the potential consequences of AAS use that apply directly to them.

The effects of an anabolic androgenic steroid and low serotonin on social and non-social behaviors in male rats.

The behavioral and neurochemical impact of low serotonin (5-HT) was examined in gonadally intact male rats exposed to an anabolic androgenic steroid (AAS) during puberty. Low 5-HT was induced beginning on postnatal day 26 using parachlorophylalanine (PCPA). Injections of the AAS, testosterone (TP), began on day 40. The rats were tested in both non-social (locomotor activity and nose poke for food) and social (low-threat and high-threat) contexts. PCPA and TP+PCPA significantly decreased locomotor activity. PCPA alone significantly increased nose poke latency compared to controls. Freezing in the PCPA group was significantly elevated compared to TP and TP+PCPA groups, but not compared to controls. AAS did not affect non-social behaviors. Thus, low serotonin may increase freezing in a non-social context. Following provocation, PCPA and TP+PCPA significantly increased aggression toward smaller non-threatening opponents, suggesting that males with low 5-HT are more aggressive in a low-threat context when provoked. In the resident-pair intruder test, TP significantly increased aggression whereas PCPA did not, suggesting that in a high-threat context, aggression is primarily mediated by AAS. TP+PCPA males were also significantly more aggressive in the high-threat context suggesting that exposure to AAS may override freezing behavior induced by low serotonin. Both PCPA and TP+PCPA significantly and substantially depleted 5-HT and 5-HIAA in all brain regions examined. AAS significantly decreased 5-HIAA levels in the hypothalamus and increased 5-HT levels in the frontal cortex. Following withdrawal from TP+PCPA, most behavioral and neurochemical measures returned to control levels. These data suggest that low serotonin may be a contributing factor in the increased aggression displayed by adolescents who abuse AAS.

Effects of anabolic androgenic steroids on the development and expression of running wheel activity and circadian rhythms in male rats.

In humans, anabolic androgenic steroid (AAS) use has been associated with hyperactivity and disruption of circadian rhythmicity. We used an animal model to determine the impact of AAS on the development and expression of circadian function. Beginning on day 68 gonadally intact male rats received testosterone, nandrolone, or stanozolol via constant release pellets for 60 days; gonadally intact controls received vehicle pellets. Wheel running was recorded in a 12:12 LD cycle and constant dim red light (RR) before and after AAS implants. Post-AAS implant, circadian activity phase, period and mean level of wheel running wheel activity were compared to baseline measures. Post-AAS phase response to a light pulse at circadian time 15 h was also tested. To determine if AAS differentially affects steroid receptor coactivator (SRC) expression we measured SRC-1 and SRC-2 protein in brain. Running wheel activity was significantly elevated by testosterone, significantly depressed by nandrolone, and unaffected by stanozolol. None of the AAS altered measures of circadian rhythmicity or phase response. While SRC-1 was unaffected by AAS exposure, SRC-2 was decreased by testosterone in the hypothalamus. Activity levels, phase of peak activity and circadian period all changed over the course of development from puberty to adulthood. Development of activity was clearly modified by AAS exposure as testosterone significantly elevated activity levels and nandrolone significantly suppressed activity relative to controls. Thus, AAS exposure differentially affects both the magnitude and direction of developmental changes in activity levels depending in part on the chemical composition of the AAS.

Behavioral effects of pubertal anabolic androgenic steroid exposure in male rats with low serotonin.

The goal of this study was to assess the interactive effects of chronic anabolic androgenic steroid (AAS) exposure and brain serotonin (5-hydroxytryptamine, 5-HT) depletion on behavior of pubertal male rats. Serotonin was depleted beginning on postnatal day 26 with parachlorophenylalanine (PCPA 100 mg/kg, every other day); controls received saline. At puberty (P40), half the PCPA-treated rats and half the saline-treated rats began treatment with testosterone (T, 5 mg/kg, 5 days/week). Behavioral measures included locomotion, irritability, copulation, partner preference, and aggression. Animals were tested for aggression in their home cage, both with and without physical provocation (mild tail pinch). Brain levels of 5-HT and its metabolite, 5-hydroxyindoleacetic acid (5-HIAA), were determined using HPLC. PCPA significantly and substantially depleted 5-HT and 5-HIAA in all brain regions examined. Chronic T treatment significantly decreased 5-HT and 5-HIAA in certain brain areas, but to a much lesser extent than PCPA. Chronic exposure to PCPA alone significantly decreased locomotor activity and increased irritability but had no effect on sexual behavior, partner preference, or aggression. T alone had no effect on locomotion, irritability, or sexual behavior but increased partner preference and aggression. The most striking effect of combining T+PCPA was a significant increase in attack frequency as well as a significant decrease in the latency to attack, particularly following physical provocation. Based on these data, it can be speculated that pubertal AAS users with low central 5-HT may be especially prone to exhibit aggressive behavior.

Factors influencing aggression toward females by male rats exposed to anabolic androgenic steroids during puberty.

Previous results showed that male rats pubertally exposed to anabolic androgenic steroids (AAS) displayed aggression towards females in response to physical provocation. This experiment examined two factors that may modulate AAS-induced behavior towards females: olfactory cues and frustration. Gonadally intact males began one of three AAS treatments at puberty (D40): testosterone propionate (T), stanozolol (S), T+S, or vehicle control. To test for the relevance of olfactory cues in the elicitation of behavior toward females, a hidden neighbor paradigm was used. The proximal stimulus was an ovariectomized (OVX) female, estrogen plus progesterone (E+P) female, or an E+P female with tape-obstructed vagina (OBS). Distal olfactory cues from a hidden neighbor were delivered from a separate cage connected to the testing arena. The vaginally obstructed, sexually receptive female (OBS) was used to determine the effects of frustration on behavior by AAS males. Both sexual and aggressive behaviors were measured. The presence of distal olfactory cues had no effect on either sexual or aggressive behavior. In the presence of E+P and OBS females, all males displayed sex behaviors, not aggression. However, AAS males displayed significantly more aggression towards proximal OVX females than controls. AAS males mounted OBS females significantly more than controls, indicating a persistence of once rewarded behavior. These results suggest (1) proximal cues of the conspecific female are more salient than distal olfactory cues in determining behavior and (2) AAS males display frustration-induced persistence in response to vaginally obstructed receptive females.

Physical provocation of pubertal anabolic androgenic steroid exposed male rats elicits aggression towards females.

Human studies suggest that anabolic androgenic steroid (AAS) users are aggressive towards women. This study used a rat model to evaluate whether AAS potentiated aggression towards females and the conditions under which this occurs. Gonadally intact pubertal male rats received one of the following AAS treatments (5 mg/kg s.c. 5 days/week for nine weeks): testosterone (T), stanozolol (S), testosterone + stanozolol (T + S), or vehicle control. Each rat was tested with 3 conspecific stimuli: ovariectomized females (OVX), estrogen only females (E), and estrogen + progesterone females (E + P). The response to physical provocation was tested under three conditions: without physical provocation, provocation of the experimental male, and provocation of the conspecific female. Provocation was a mild tail pinch. Both aggressive and sexual behaviors were measured during each test. In the absence of physical provocation, AAS males were not aggressive towards females. However, provocation significantly increased aggression in males treated with testosterone but only towards OVX females. In the presence of E or E + P females, all animals displayed sex behavior, not aggression. Thus, factors such as the nature of the AAS and the hormonal status of the females are important in determining whether male rats will be aggressive towards females. However, the most salient factor determining aggression towards females is the presence of provocation in combination with high levels of testosterone.

Stacking anabolic androgenic steroids (AAS) during puberty in rats: a neuroendocrine and behavioral assessment.

Anabolic androgenic steroid (AAS) abuse is increasing in teenagers. We examined the effects of stacked AAS in adolescent male rats. Stacking, in which multiple AAS are taken simultaneously, is commonly employed by humans. Beginning at puberty gonadally intact male rats received testosterone, nandrolone, or stanozolol. Additional groups received stacked AAS: testosterone + stanozolol, nandrolone + stanozolol, or nandrolone + testosterone. Injections continued during tests for sexual behavior, vocalizations, scent marking, partner preference, aggression and fertility. Body and reproductive tissue weights were taken. Sexual and aggressive behaviors were increased by testosterone yet inhibited by stanozolol; nandrolone had no effect. Stacking testosterone with stanozolol prevented the inhibitory effects of stanozolol. Body weight was decreased by testosterone and all stacked AAS. Cell nuclear androgen receptor binding in brain was significantly increased in nandrolone males and decreased in stanozolol males; testosterone males were slightly higher than controls. Androgen receptors in stacked groups were intermediate between individual AAS suggesting that stanozolol competed with other AAS for androgen receptors despite its low affinity. The results indicate that stacking AAS influences the effects of individual AAS on behavioral and endocrine measures, and levels of androgen receptor occupation are not directly correlated with AAS effects on behavior.

Microlesions of the ventromedial nucleus of the hypothalamus: effects on sociosexual behaviors in male rats.

Electrolytic microlesions aimed at the dorsomedial portion of the ventromedial nucleus (VMN) of the hypothalamus were generated, and effects on copulation, 50-kHz vocalizations, scent marking, and sexual motivation were measured. Male rats were tested before and after lesions, after castration, and after testosterone replacement. Three control groups were used: One received sham surgery, another received no surgery or testosterone replacement, and a 3rd received lesions primarily outside the VMN. VMN lesions produced impairments in testosterone's ability to restore ultrasonic vocalizations and scent marking, assessed with 2 different test methods. Copulation, sexual motivation, and weight gain were largely unaffected, although some differences were observed in copulatory efficiency. The authors conclude that the integrity of the VMN is important for full expression of sociosexual behaviors in male rats.

Long-term effects of pubertal anabolic-androgenic steroid exposure on reproductive and aggressive behaviors in male rats.

The current study examined acute and long-term effects of anabolic-androgenic steroid (AAS) exposure during puberty on copulation, vocalizations, scent marking, and intermale aggression, both with and without tail pinch, in intact male rats. Animals received 5 mg/kg of testosterone, nandrolone, stanozolol, or vehicle, beginning at puberty. After 5 weeks, behavior tests were performed while continuing AAS injections. AAS treatment was then discontinued. Behaviors were tested during 3-5 weeks, 9-11 weeks, and 15-17 weeks of withdrawal. During AAS administration, stanozolol males showed significant reductions in all behaviors compared with controls, except aggression with tail pinch. Nandrolone treatment significantly reduced vocalizations and scent marking, and testosterone had no significant effect on behavior. During withdrawal, behaviors in stanozolol males recovered to control levels at variable rates: aggression at 4 weeks; mounts, vocalizations, and scent marking at 9 weeks; and ejaculations at 15 weeks of withdrawal. Stanozolol males showed significantly higher levels of tail pinch-induced aggression during every withdrawal test. Nandrolone-treated males scent-marked at control levels by 9 weeks withdrawal but displayed significantly fewer vocalizations and significantly more tail pinch-induced aggression than controls for the entire study. Testosterone-treated males scent-marked significantly below controls at 3 weeks withdrawal and showed significantly more tail pinch-induced aggression at 5 weeks withdrawal. All three AAS significantly increased tail pinch-induced aggression compared with corresponding nontail pinch tests, even at study endpoint. These results suggest that alterations in androgen-dependent behaviors by pubertal AAS exposure can persist long after drug exposure, and some effects may even be permanent.

Androgen receptor blockade in the MPOA or VMN: effects on male sociosexual behaviors.

Previously, our laboratory has shown that androgen receptors in the medial preoptic area (MPOA) and ventromedial nucleus (VMN) are necessary for copulation in male rats. The present study examined whether these receptors are required for other sociosexual behaviors. In Experiment 1, different regions of the VMN were implanted with the antiandrogen hydroxyflutamide (OHF). We found that implants located in anterodorsal portions of the VMN were more effective at inhibiting the restoration of copulation than implants in the posteroventral VMN. In Experiment 2, a second set of male rats was pretested for copulation and other sociosexual behaviors and was castrated. Experimental animals then received Silastic capsules filled with testosterone (T) plus intracranial (IC) implants filled with OHF to selectively block androgen receptors in either the MPOA or VMN. We found that androgen receptor blockade in the MPOA inhibited the restoration of copulation but had no effect on other sociosexual behaviors. OHF directed at the VMN inhibited the restoration of copulation and 50-kHz vocalizations but had no effect on scent marking. Two tests were used to assay sexual motivation: partner preference and conditioned place preference (CPP). Both methods revealed impairments in sexual motivation in the VMN group but not in animals receiving OHF in the MPOA. Taken together, these data suggest that androgen receptors in the MPOA are essential for copulatory performance, while androgen receptors in the VMN are important for copulation, sexual motivation, and androgen-dependent vocalizations.

Anabolic androgenic steroids and aggression: studies using animal models.

The use of anabolic androgenic steroids (AASs) has escalated in teenagers and is associated with increased violence. Adolescent exposure to chronic high levels of AASs is of particular concern because puberty is a hormonally sensitive period during which neural circuitry for adult male patterns of behavior develop. Thus, teenage AAS use may have long-term repercussions on the potential for displaying aggression and violence. Animal models have contributed valuable information on the effects of AAS use. For example, studies in rodents confirmed that exposure to the AASs testosterone and nandrolone, but not stanozolol, does indeed increase aggression. A side effect of AAS use reported in humans is "'roid rage," characterized by indiscriminate and unprovoked aggression. Results of animal studies demonstrated that pubertal rats receiving AASs respond appropriately to social cues as they are more aggressive toward intact males than are castrates. Also, testosterone-treated males recognize appropriate environmental cues as they are most aggressive in their home cage. Thus, adolescent AAS exposure increases aggressive behaviors, but does not induce indiscriminate aggression. To assess whether AAS exposure increases aggression after provocation, rats were tested following a mild tail-pinch. In adolescent males, provocation increased aggression after withdrawal from testosterone, nandrolone, and stanozolol, an effect which persisted for many weeks. The data suggest that AASs sensitize animals to their surroundings and lower the threshold to respond to provocation with aggression. Thus, in humans, pubertal AAS exposure may not cause violent behaviors, but may increase the likelihood that aggressive acts will result in violence. This may persist into adulthood.

Effects of pubertal anabolic-androgenic steroid (AAS) administration on reproductive and aggressive behaviors in male rats.

Adolescence in human males is a hormonally sensitive period when many adult behaviors develop, including sexual and aggressive behaviors. Using a rat model, the authors examined the effects of three anabolic-androgenic steroids (AAS) during puberty: testosterone, nandrolone, and stanozolol. Copulation, vocalizations, scent-marking, and aggression were tested following AAS exposure. Relative to gonadally intact controls, rats injected with testosterone showed a significant increase in scent-marking and aggression in the opponent's home cage. Nandrolone had no effect. Stanozolol significantly inhibited all behaviors. Results suggest that depending on the chemical structure of the steroid, AAS exposure during puberty affects several androgen-dependent behaviors. Because adolescence in humans is a period of hormonal change, abuse of AAS, particularly stanozolol, during this time may disrupt the establishment of normal adult behavior patterns.

Characterization of 50-kHz ultrasonic vocalizations in male and female rats.

Male and female rats emit ultrasonic vocalizations in reproductive encounters. While estrous bedding has been used to elicit vocalizations of males, the number of responses is variable. We report a reliable method to assess vocalizations using exposure to a stimulus animal. The stimulus rat is placed behind a wire barrier for 5 min, then removed. Vocalizations are then recorded for 5 min. Experiment 1 validated this method and it was used for subsequent experiments. In Experiment 2, male rats were castrated and tested for the restoration of vocalizations. In one group, males were allowed to copulate freely; in the other, females had vaginal masks to prevent ejaculation, but not mounting. Vocalizations were restored only in males allowed to ejaculate. In Experiment 3, we measured vocalizations in sexually nai;ve and sexually experienced males following exposure to either castrated (CAS) males, testosterone (T)-treated males, ovariectomized (OVX) females, or OVX females receiving estrogen plus progesterone (E+P). Males vocalized most after exposure to E+P females, whether they were sexually experienced or naive. However, the rate of vocalizations was significantly higher after exposure to E+P females when the males were sexually experienced. In Experiment 4, we measured vocalizations in females following exposure to CAS males, T-treated males, OVX females, or E+P females. Females vocalized most after exposure to T-treated males. Our results show that (1) sexual experience facilitates vocalizations in male rats, (2) vocalizations are highest after exposure to hormonally receptive conspecifics, and (3) ultrasonic signaling is a sensitive index for assessing the hormonal disposition of conspecifics.

Effects of testosterone in the VMN on copulation, partner preference, and vocalizations in male rats.

The present study tested whether testosterone propionate (TP) implanted in the ventromedial nucleus (VMN) of the hypothalamus could initiate performance, motivational, or sociosexual components of sexual behavior in castrated male rats. Twenty-seven intact male Long Evans rats were pretested for copulation, partner preference, and 50-kHz vocalization and were subsequently castrated. Approximately 3 weeks after castration, males were retested to confirm that these behaviors had declined, and groups were assigned. Groups 1 and 2 were implanted with bilateral stainless steel cannulae directed at the VMN that were either filled with TP (TVMN group) or remained empty (Blank group). A third group (TSC) was implanted subcutaneously with two 10-mm Silastic capsules filled with testosterone. Restoration of behavior was measured for 2 weeks after implants. We found that copulation and 50-kHz vocalization were not restored by TP in the VMN alone. However, partner preference returned to preoperative levels in both the TVMN and TSC groups, indicating that TP in the VMN was sufficient to restore sexual motivation. Following behavioral testing, prostate glands and seminal vesicles were weighed and confirmed that TP did not leak into the periphery in the TVMN group. Immunostaining for androgen receptors also verified that TP spread was confined to the immediate area surrounding the cannula tip. These results suggest that androgen activation at the VMN is sufficient to induce the motivational components of male sexual behavior, whereas activation of other brain sites is required for copulation and ultrasonic vocalization.

Effects of hydroxyflutamide in the medial preoptic area or lateral septum on reproductive behaviors in male rats.

We examined whether androgen receptors in the medial preoptic area (MPOA) and lateral septum (LS) are required for the expression of copulation and sexual motivation. Castrated males received testosterone-filled silastic capsules to restore behavior, and were implanted with the antiandrogen hydroxyflutamide (OHF) or blank cannulae. One group was implanted in either the anteroventral MPOA or LS (ANT group). Another group was implanted in the posterodorsal MPOA or LS (POST group). Copulation was tested on days 2, 6, 10, and 14 of OHF exposure; partner preference, a measure of sexual motivation, was tested on day 15. The results showed that sexual behavior was significantly suppressed by OHF in the MPOA of the ANT group, but not the POST group. However, sexual motivation was significantly reduced by OHF in the MPOA of the POST group, but not the ANT group. In the LS, OHF had no effect on sexual behavior and partner preference regardless of implant site. The data suggest site specificity within the MPOA for androgen receptor activation of male reproductive behaviors.

Effects of withdrawal from anabolic androgenic steroids on aggression in adult male rats.

In gonadally intact male rats, chronic exposure to high levels of testosterone propionate (TP) increases aggression, nandrolone (ND) has little effect and stanozolol (ST) suppresses aggression. The present experiment tested whether the effects of TP, ND and ST on aggression and reproductive tissues are reversed following anabolic androgenic steroid (AAS) withdrawal. Gonadally intact males received TP, ND, ST or vehicle for 12 weeks. Injections were then discontinued. Aggression was tested 3 weeks (short term) and 12 weeks (long term) after withdrawal of AAS treatment, with either a gonadally intact or a castrated opponent in three different environments (home, opponent's and neutral cage). After short-term withdrawal, some parameters of aggression were significantly above control levels in TP males. There were no significant differences between ND or ST males and controls, though ST males showed the lowest levels of aggression. No significant differences between any of the groups were found after long-term withdrawal. Eighteen weeks after AAS withdrawal, serum testosterone (T) and LH levels were comparable to controls in all groups. Testes weights were at control levels in ST males, but significantly higher than controls in TP and ND males. Seminal vesicle weights were significantly elevated in TP males, but similar to controls in both ND and ST males. None of the prostate weights were significantly different from controls. These results suggest that aggression gradually returns to normal following withdrawal from AAS. Some, if not all, hormone levels and tissue weights return to normal, suggesting possible long-lasting effects of chronic AAS exposure.