Hormonal Pathways Regulating Intermale And Interfemale Aggression

HORMONAL PATHWAYS REGULATING INTERMALE AND INTERFEMALE AGGRESSION

Neal G. Simon, Qianxing Mo, Shan Hu, Carrie Garippa, and Shi-fang Lu Department of Biological Sciences, Lehigh University, Bethlehem, Pennsylvania 18015, USA

I. Introduction A . Common Regulatory Concepts in Males and Females II. Females A . DHEA as a Neurosteroid III. Males A . Regulation in the Adult B . Neural Steroid Receptors IV. Hormonal Modulation of Serotonin Function V. Conclusions References

I. Introduction

Characterization of the mechanisms involved in the regulation of aggression by androgens is a major objective in behavioral endocrinology. Advances in our understanding of molecular, cellular, and biochemical processes that mediate androgenic eVects in target cells (Lee and Chang, 2003) continue to drive revisions in increasingly sophisticated models of behavioral regulation in animal models. Investigations that seek to discern hormonal contributions to aggression in clinical populations have not seen comparable progress. These studies face significant methodological limitations that, combined with wide variation in indices of aggression, frequently lead to equivocal results (Archer, 1991, 1988; van der Pahlen, 2005). As subtypes of aggression, such as irritability, impulsivity, hostility, and dominance are employed as target behaviors versus a global construct termed ‘‘aggression’’ in human studies, advances in defining the relationship between hormones and specific behavioral forms should emerge (Simon, 2002). OVensive aggression between conspecific males and conspecific females can serve as model systems to exemplify our understanding of androgenic eVects on aggressive behavior. This form of aggression between same-sex conspecifics is a productive behavior because it determines dominance status and access to resources. The use of oVensive aggression in males as a model is based on its widely documented dependence on testosterone (T), the principal testicular INTERNATIONAL REVIEW OF NEUROBIOLOGY, VOL. 73 DOI: 10.1016/S0074-7742(06)73003-3

99

Copyright 2006, Elsevier Inc. All rights reserved. 0074-7742/06 $35.00

100

SIMON et al.

androgen (Nelson, 1995). For well over 100 years, it has been recognized that gonadectomy reduces conspecific aggression in males (Freeman et al., 2001). Including females as a representative system for exemplifying androgenic eVects on aggression may seem unusual given the numerous failures to identify a positive relationship between T and this behavior in female mammals. However, several studies clearly demonstrated that females housed in small groups displayed aggression toward other females, juvenile males, or gonadectomized adult males (Brain and Haug, 1992) and that dehydroepiandrosterone (DHEA), an androgenic neurosteroid synthesized in the brains of humans and other mammals (Baulieu et al., 2001; Compagnone and Mellon, 2000), played an important role in regulating this behavior. Assessments of seasonal variation in aggression in avian species supported the concept that DHEA also may influence the display of male-typical aggression, particularly outside the breeding season (Hau et al., 2004; Soma and Wingfield, 2001). A systems perspective has been adopted in our laboratory to frame the relationship between androgens and conspecific oVensive aggression in males and females (Fig. 1). This approach draws on recent developments in functional genomics, cell biology, biochemistry, and molecular biology to build hypotheses and develop regulatory models that span gene function through behavioral expression. Environmental influences on behavior and adaptive responses to these events are additional important features of the systems approach. This aspect of the model recognizes the influence of factors, such as age, cognition, experience, diet, and culture on signaling pathways.

FIG. 1. A graphic outline of the components that require characterization for the development of a system’s model for the regulation of conspecific oVensive aggression. Progress in developing the model will require a multidisciplinary approach that draws on molecular and cell biology, bioinformatics, physiology, ethology, ecology, and evolutionary biology. The application of a systems analysis to the relationship between androgens and oVensive aggression should yield a model that integrates events from the gene level to behavioral expression and subsequent adaptations based on experience. Adapted from Simon and Lu, in press, and reprinted with permission from Oxford Press.

ANDROGENS AND AGGRESSION

A. COMMON REGULATORY CONCEPTS

IN

MALES

AND

101

FEMALES

1. Metabolism Aromatization is widely recognized as an important step in the promotion of aggression by T (Balthazart et al., 2003). Evidence has also accumulated showing that androgens can directly induce male-typical aggression (Simon, 2002). These observations demonstrate that defining the role of T in male-typical, oVensive aggression must include a discussion of the contributions of E2 and dihydrotestosterone (DHT), the metabolites produced by the activity of aromatase and 5
reductase, respectively. The distribution of these enzymes in the CNS (Melcangi et al., 1998; Naftolin et al., 2001; Silverin et al., 2000) and their localization within sites implicated in male-typical aggression are important considerations. Several methods and strategies have been employed to assess the eVects of these metabolites. Among the most common are behavioral assessments in mice with disruptions of specific steroid receptor genes or key enzymes (ER
, ER, aromatase), descriptions of aggressive phenotypes in mice with naturally occurring mutations that aVect receptor function or critical enzyme activity (e.g., Tfm, 5
-reductase deficient), pharmacological manipulations (enzymatic inhibitors, antagonists), and comparisons among outbred strains in the postcastration response to specifically acting androgenic and estrogenic hormones. The modulatory actions of DHEA on female-typical aggression may involve multiple metabolites of this steroid, a circumstance that would parallel observations in males. The synthetic and metabolic pathways for DHEA have been well defined (Compagnone and Mellon, 2000; Labrie, 2003) (Fig. 2). For aggression, the 3-hydroxysteroid dehydrogenase (3-HSD), hydroxysteroid sulfotransferase (HST), steroid sulfatase (SST), and CYP7B pathways all may be involved. The activity of 3-HSD has attracted substantial interest because it leads to androstenedione (AE) formation, which can serve as substrate for the production of more potent androgens and estrogens. The balance between SST and HST may determine the contribution of DHEA sulfate (DHEA-S) versus DHEA, and CYP7B family activity leads to the production of 7
- and 7-hydroxy DHEA. The activity of CYP7B enzymes has been neglected in the context of female-typical aggression, which may represent a significant gap in the literature because the 7
hydroxylated metabolite of DHEA appears to be the major form recovered in the CNS (Cui and Belshams, 2003; Jellinck et al., 2001, 2005). Further complicating characterization of the mechanisms through which DHEA aVects aggression in females are data showing that both genomic and nongenomic eVects may be involved. Direct androgenic eVects of DHEA itself, and the observation that more potent androgens are formed from DHEA in peripheral tissues (Labrie, 2003; Lu et al., 2003; Mo et al., 2004) have provided evidence that establishes genomic activity. In relation to nongenomic eVects of DHEA, metabolism is important

102

SIMON et al.

FIG. 2. A summary of the metabolism of DHEA in the central nervous system. Three pathways have been identified with DHEA as the initial substrate: (A) to DHEA sulfate, a reversible path involving hydroxysteroid sulfotransferase and steroid sulfatase, (B) to 7
- or 7-hydroxy DHEA, which involves CYP7B pathways, and (C) to androstenedione, which utilizes 3-hydroxysteroid dehydrogenase and provides the possibility for the formation of more potent androgens and estrogens. Adapted from Simon (2002) and reprinted with permission from Academic Press.

because there are diVerences in the potency of DHEA versus DHEA-S as negative modulators of the GABAA receptor (Imamura and Prasad, 1998). 2. Neuromodulator Hypothesis A conceptual model that enables the integration of androgenic eVects on aggression in males and females would have broad utility. If T and DHEA are seen as modulators of neurochemical systems, it becomes possible to propose a model that bridges observed eVects in both sexes. We have termed this model the

ANDROGENS AND AGGRESSION

103

neuromodulator hypothesis. The strength of this conceptual approach is a strong emphasis on integration among hormonal, neurochemical, and peptide systems that influence aggression and, if validated, the capacity to bridge findings about aggression in animal models with clinical issues related to androgen excess or deficiency. To illustrate the utility of the neuromodulator hypothesis, the influence of T and DHEA on representative neurochemical systems is presented in both females and males. II. Females

A widely held premise of eVorts to understand the hormonal contribution to female-typical aggression has been that aggressive behavior exhibited by females follows the same regulatory processes observed in males, that is, an emphasis on a facilitative contribution of T. Not surprisingly, when this position has been tested experimentally, mixed to outrightly negative outcomes have been obtained (Albert et al., 1993; Giammanco et al., 2005; von Engelhardt et al., 2000). We believe that a diVerent conceptual approach to hormone function in femaletypical oVensive aggression is needed. Based on over 20 years of findings, which show that the neurosteroid DHEA inhibits female-typical aggression when administered chronically (Baulieu, et al., 2001; Young et al., 1995, 1996), a model that focuses on the eVects and mechanism of action of this compound may have utility. Intact or ovariectomized females reliably display attack behavior toward intruder females that are intact, ovariectomized, or lactating (Brain and Haug, 1992; Simon, 2002). This type of aggression appears to be under GABAergic control and is modulated by DHEA (Simon, 2002; Young et al., 1995, 1996), a neurosteroid synthesized in the CNS (Baulieu et al., 2001; Compagnone and Mellon, 2000). The demonstration that extended treatment with DHEA inhibited aggression by intact or OVX females toward females or lactating females generated interest in this steroid. In addition to modulating GABA function, DHEA exerts androgenic eVects through the androgen receptor and also may serve as substrate for more potent steroidal metabolites (Fig. 2). These recent findings raise the possibility that the eVects of DHEA on female-typical aggression are exerted through multiple mechanisms. A. DHEA

AS A

NEUROSTEROID

DHEA modulates GABAA, NMDA, and 1 receptors (Compagnone and Mellon, 2000), although eVects exerted at the GABAA receptor complex have received the most attention (Dubrovsky, 2005; Rupprecht, 2003; Rupprecht et al.,

104

SIMON et al.

2001). The emphasis on GABAA receptors in female-typical aggression is supported by numerous studies, which showed specific eVects of GABA on oVensive aggression (Miczek et al., 2003; Siegel et al., 1999). Also, the NMDA receptor, to the extent it has been studied in relation to aggression (Adamec, 1997; Blanchard et al., 2002; Gould and Cameron, 1997) is linked to defensive behavior. Last, to the best of our knowledge, the 1 receptor has little influence on aggression, although an indirect role cannot be excluded. This is because 1 receptor can eVect NMDA-mediated responses (Maurice et al., 1999). 1. Mechanism of Action The prevailing position concerning the modulation of female-typical aggression by DHEA is that it produces a reduction in pregnenolone sulfate (Preg-S), a potent negative modulator of the GABAA receptor (Majewska and Schwartz, 1987), through competition for HST. By decreasing Preg-S, GABA function is enhanced, which in turn inhibits oVensive aggression (Robel and Baulieu, 1995). DHEA also can act at membrane sites rapidly to alter receptor conformation (a nongenomic eVect) and aVect long-term processes by itself or through neurosteroid metabolites (a genomic eVect). The resulting changes in gene expression could then alter membrane receptor function by, for example, producing changes in GABAA subunit composition (Herbison and Fenelon, 1995). An important feature of GABAA receptors is that modulation can be achieved at multiple sites because of its pentameric structure. Included are sites that bind GABA, the benzodiazepines, the C1
ionophore, barbiturates, and an as yet unidentified neurosteroid binding site (Majewska, 1995; Majewska and Schwartz, 1987; Majewska et al., 1990). For female-typical aggression, however, prevailing models have focused on the reduction in Preg-S as a critical eVect rather than a direct action on GABAA receptor. Interpreting the eVects of DHEA on aggression is not a straightforward proposition. DHEA itself is a negative modulator of GABAA receptor (albeit weaker than Preg-S) and a positive modulator of NMDA receptor (Bergeron et al., 1996; Imamura and Prasad, 1998; Majewska, 1992; Sousa and Ticku, 1997). A positive association between aggression and DHEA levels was reported in some avian species and adolescents with conduct disorder, which suggests a potential facilitative eVect (Hau et al., 2004; Soma and Wingfield, 2001; van Goozen et al., 2000). The aggression-enhancing eVects, however, were noted in males and, in these circumstances, likely reflect direct androgenic eVects of DHEA (Lu et al., 2003; Mo et al., 2004). The recent finding of direct androgenic eVects of DHEA may provide new insights into underlying mechanisms. DHEA exhibited characteristics of typical androgenic compounds, which included upregulation of androgen receptor (AR) protein expression (Fig. 3) and conferring AR transcriptional activity in a dosedependent manner (Lu et al., 2003; Mo et al., 2004). Western analysis of brain

ANDROGENS AND AGGRESSION

105

FIG. 3. Upregulation of AR by DHEA in mouse brain. Female mice were ovariectomized one week before being treated with 80, 320, or 1280 mg or vehicle (n ¼ 3) for four consecutive days (s. c.). Five hours after the last injection, major limbic system regions were blocked and analyzed by Western blot for relative concentration of AR (Lu et al., 1998). DHEA treatment augmented cellular AR level in a dosedependent manner. A trend analysis showed the eVect of DHEA dosage was significant ( p < 0.01). Data shown are mean integrated band densities (IBD þ SEM) for 97-kD AR bands. *; significantly diVerent from controls ( p < 0.05). Adapted from Lu et al. (2003) and reprinted with permission.

extracts from LS, BNST, and MPOA showed a dose-dependent increase in AR content in response to DHEA treatment, and a similar regulatory eVect also was seen in GT1-7 cells, which are AR-expressing hypothalamic cell lines. Importantly, the upregulation of AR by DHEA was not blocked by trilostane, an inhibitor of 3-hydroxysteroid dehydrogenase activity responsible for the conversion of DHEA to androstenedione, a more potent androgen. The androgenic activity of DHEA was further confirmed when it was shown that DHEA induced intracellular translocation of AR-GFP and formation of nuclear clusters (Mo et al., 2006). When COS-7 cells transfected with an ARGFP expression vector were treated with 10
7 M DHEA for 24 hours, AR-GFP protein translocated from the cytoplasm into the nucleus and led to the formation of punctate fluorescent foci (Fig. 4, Mo et al., 2006). The androgenic activity of DHEA represents a novel mechanistic finding that also may be a potential component of the antiaggressive mechanism. Another mechanistic component tied to genomic eVects of DHEA or its metabolites is an alteration of GABAA subunit structure, which potentially can influence the extent of modulation (Herbison and Fenelon, 1995; Mehta and Ticku, 1999). The direct androgenic eVects of DHEA and its metabolites thus

106

SIMON et al.

FIG. 4. AR-GFP protein intracellular traslocation and nuclear clustering in response to DHEA or control treatment. COS-7 cells were transfected with AR-GFP expression plasmid pEGFP-N1-AR. AR-GFP fusion proteins were detected in living cells by excitation with 488 nm line from an argon laser of a Zeiss LSM-510 confocal microscope. (A) Typical COS-7 cells without treatment. (B) COS-7 cells were treated with 10
7 M DHEA for 24 hours. Bar ¼ 5 mm.

may represent a crosstalk cellular signaling system (Katzenellenbogen, 2000; Rupprecht, 2003) linked to its antiaggressive eVect. It is critically important to define the interrelationship among DHEA, its androgenic eVects, the subunit structure of GABAA receptor, and attendant changes in function to elaborate the mechanism of action of this neurosteroid and how it modulates the expression of female-typical aggression. Establishing the functional significance of observed alterations in GABAA structure, whether produced directly at the membrane level or through the AR, requires additional steps. A neuroanatomy of female-typical aggression needs to be defined. This is particularly important because modulation of GABAA

ANDROGENS AND AGGRESSION

107

receptor by DHEA can occur through multiple mechanisms that diVer across regions. For example, GABAA receptor subunit structure varies regionally and is a limiting factor in steroidal eVects (Mehta and Ticku, 1999). In addition, AR distribution and the GABA system only partially overlap. Thus, androgenic activity of DHEA may play an important role in some regions, while nongenomic eVects of DHEA are essential elsewhere.

III. Males

The ability of T to facilitate the display of intermale aggressive behavior is recognized as a fundamental relationship in behavioral endocrinology. Over the past 30 years, the focus of research in the field has shifted largely to mechanistic questions that addressed neuroanatomical, cellular, and molecular processes involved in hormonal responsiveness. Rodent models have been a primary tool in these investigations and their utility for providing data directly relevant to humans has been buttressed by genomic data and molecular conservation of steroidal systems in mammals (Choong et al., 1998). Animal models have broadened considerably to include species ranging from fish to lizards and, increasingly, birds (Elofsson et al., 2000; Godwin and Crews, 2002; Panksepp, 2003; Wingfield et al., 1997; Woolley et al., 2004). The overarching goal of these investigations has been to characterize CNS pathways in the adult brain that underlie the ability of T to promote aggressive behavior. Comparisons of sex and strain diVerences in the response to T and its major metabolites, E2 and DHT, as well as studies using enzymatic inhibitors and receptor antagonists, were important steps in elaborating these pathways (Simon, 2002). Models of steroid receptor function and cellular mechanisms involved in the hormonal regulation of aggression have grown increasingly sophisticated. The complexity of these models has resulted from eVorts to carefully define the molecular processes that determine cellular sensitivity to the aggressionpromoting property of gonadal steroids. Achieving a comprehensive regulatory model may well require more than an understanding of hormonal systems in isolation; elaboration of interactions between steroidal and relevant neurochemical systems will be needed. At present, the largest amount of data in this area is in regard to hormonal influences on serotonin function in males. For females, the modulation of GABAA receptor function by DHEA has been a major focus. In the following sections, hormonal influences on these target neurochemical systems will be used to exemplify the utility of the neuromodulator hypothesis.

108

SIMON et al.

A. REGULATION

IN THE

ADULT

The metabolism of T peripherally and in CNS target cells established a physiological basis for multiple steroidal pathways regulating aggression in males (Simon, 2002). Four distinct steroid-sensitive pathways have been identified: (i) Androgen-sensitive, which responds to T itself or its 5
-reduced metabolite, DHT (ii) Estrogen-sensitive, which uses E2 derived by aromatization of T (iii) Synergistic or combined, in which both the androgenic and estrogenic metabolites of T are used to facilitate behavioral expression (iv) Direct T-mediated, which utilizes T itself It is important to note that a given male does not necessarily express all four systems; genotype is the major determinant of the functional pathway. Estrogen is the most typical active hormone, which is consistent with a prominent role for aromatization and estrogen receptor. The regulatory pathways share a basic feature of high sensitivity in males. It takes only 2–3 days of hormone treatment with the appropriate steroid at physiological doses to restore aggression, a time course in keeping with a genomic mechanism of action.

B. NEURAL STEROID RECEPTORS The characterization of multiple neuroendocrine pathways through which T can facilitate aggressive behavior provided a basis for assessing the contributions of androgen receptor (AR) and estrogen receptor (ER) in these systems. The time frame for postcastration restoration of aggressive behavior in mice and other rodents strongly supports a focus on classical genomic processes, that is, on these receptors as transcription factors. Recent developments have added support to the critical role of AR (Sato et al., 2004), and the role of ER subtypes has been more clearly defined (PfaV et al., 2002). 1. Androgen Receptor The importance of AR in male typical aggression has not been well appreciated, in part due to an emphasis on aromatization, the formation of E2, and subsequent activation of ER-mediated signaling pathways (Balthazart et al., 2003; PfaV et al., 2002). The importance of AR in the expression of aggressive behavior was reinforced by results that showed that AR gene knockout (ARKO) in male mice led to the ablation of male-typical aggressive behavior (Sato et al., 2004). Further, the impaired male typical behavior in female ER
knockout (ERKO) mice was restored by DHT treatment. Developmental experiments revealed that

ANDROGENS AND AGGRESSION

109

perinatal DHT treatment of female ERKO mice established sensitivity to both E2 and DHT for the induction of male-typical behaviors, but that brain masculinization was abolished upon AR inactivation. These findings indicate that: (1) AR, as well as ER, is required for the expression of male typical behaviors in both sexes; (2) enhanced activation in an androgenic signaling pathway is adequate to compensate for the loss of ER function; and (3) AR plays an organizational role in brain masculinization during development. Specific contributions of the androgenic and estrogenic metabolites of T in masculinization of the regulatory pathways for intermale aggression had been reported previously (Simon et al., 1996). Several regions that are part of the neuroanatomical substrate for conspecific aggression, including the bed nucleus of the stria terminalis (BNST), lateral septum (LS), medial preoptic area (MPOA), and medial amygdala (MAMYG), exhibit strong positive immunoreactivity for AR in rodents and nonhuman primates (Lu et al., 1998, 2006). These descriptive findings are useful for defining functional circuitry and can help elucidate the androgenic signaling cascading that eventually influences behavior. For example, a robust feature of androgen action in target cells, including the brain, is the autoregulation of AR by its cognate hormones. The characterization of sex, genotypic, or regional diVerences in androgen-induced AR autoregulation represent possible mechanisms that could underlie variation in sensitivity to the aggression-promotion property of androgen. The postcastration regulation of AR by diVerent androgens and estrogens in multiple brain regions and species has been tested. In CF-1 male and female mice, for example, a strain that is highly responsive to direct androgenic stimulation (Simon, 2002), AR regulation was dose- and ligand-dependent in multiple brain regions. Castration led to a rapid and pronounced loss of AR immunoreactivity, and T replacement (50–1000 mg) produced a dose-dependent linear increase in AR protein (Lu et al., 1998). Further, DHT, which is a more potent androgen than T, produced greater upregulation for a more extended period of time. The findings in mice recently were extended to a nonhuman primate model with similar results (Lu et al., 2006). Male cynomolgus monkeys were gonadectomized and treated with silastic implants containing E2 or DHT; control males were sham operated. The results (Fig. 5) showed that GDX þ DHT males exhibited the strongest AR immunoreactivity in the hypothalamus, while AR protein expression in GDX þ E2 males was significantly lower than controls. Identical AR regulatory processes in female mouse brain to those seen in males indicates that a rapid increase in AR protein is only one component of the processes mediating responsiveness to the aggression-promoting property of androgen. Supporting this position are repeated demonstrations that the induction of male-typical aggression in ovariectomized females requires 16–21 days of androgen treatment (Simon, 2002). Because AR level can be increased dramatically within 3 hours of androgen administration, it is likely that increased

110

SIMON et al.

FIG. 5. EVects of steroid hormones on AR content in ventromedial hypothalamus of male cynomolgus monkey brain. Animals were either sham operated or gonadectomized (GDX) and given silastic implants containing vehicle, dihydrotestosterone, or estradiol. The treatment groups were: sham þ vehicle; GDX þ vehicle; GDX þ estradiol (E2); and, GDX þ dihydrotestosterone (DHT). After 12 weeks of treatment, brains were immediately frozen at necropsy and stored at
70  C until use. Brain regions were isolated, fixed in 3.7% formaldehyde/PBS for 24 hours, sections frozen on a microtome, and processed for AR immunohistochemical (IHC) staning. Representative AR IHC staining from VMH in each of the four groups is shown in 2(A), and 2(B) provides a graphical representation of semiquantitative image analysis, * indicates significant diVerence from GDX group. Bar ¼ 50 mm. From Lu et al., 2006.

ANDROGENS AND AGGRESSION

111

cellular AR triggers enhanced (or suppressed) transcription of other androgenregulated genes. A reasonable hypothesis is that the changes in gene function bring about alterations in neuronal structure and neurotransmitter function that enables the expression of aggressive behavior. More specifically, the time frame required to induce male-like aggression in females raises the possibility that AR mediates elaboration of an androgen-dependent circuit through interactions with growth factors (Bimonte-Nelson et al., 2003; Fusani et al., 2003; Yang and Arnold, 2000; Yang et al., 2004). The pronounced sexual dimorphisms in neural pathways mediating reproductive behaviors are consistent with androgen-mediated circuit remodeling (Hutton et al., 1998; Simerly, 1998). Several of these structures, including the vomeronasal organ, accessory olfactory bulbs, medial and posterior nuclei of the amygdala, and BNST, process pheromonal and other olfactory stimuli (Segovia and Guillamon, 1993; Simerly, 1998). Because intermale aggression is triggered by a pheromonal stimulus, androgenic stimulation may function to establish this pathway in females and maintain it in normal males. AR-induced circuit remodeling in mammals may be similar to a testosterone-dependent increase in BDNF in adult male canary brain, which seems to play an important role in the viability of high vocal center neurons (Rasika et al., 1999). 2. Estrogen Receptor Elucidating the potential role of ER in the regulation of aggression became a more formidable challenge when ER, a novel form, was cloned from a variety of species including rat and human (Koehler et al., 2005; Matthews and Gustafson 2003; Sierens, 2004; Wilkinson et al., 2002). The
and  ER subtypes are highly conserved across species and share significant amino acid sequence homology, particularly in the DNA-binding and ligand-binding domains (Ogawa et al., 1998). However, ER diVers from ER
in two important aspects: in relative tissue distribution and cellular localization within the CNS (Shughrue et al., 1997, 1998) and in the relative aYnity of both naturally occurring and synthetic ligands (Kuiper et al., 1997; Sun et al., 1999). Both receptor subtypes are involved in mediating the eVects of estrogen on male-typical aggression, but their respective actions diVer markedly. Studies in estrogen receptor knockout mice (ERKO) have demonstrated that ER
is the primary facilitator of oVensive aggression (PfaV et al., 2002). In the residentintruder paradigm, oVensive attacks were rarely displayed by
ERKO males while wild-type (WT) and heterozygous males showed normal attack durations. Castration-hormone replacement studies built on these observations by showing that daily TP injections were ineVective in promoting aggression in
ERKO males but highly eVective in gonadectomized WT males. Results with ER knockout males (ERKO) provided additional support for the facilitative role for ER
because ERKO males exhibited normal or enhanced attack behavior compared to WT males.

112

SIMON et al.

ER appears to be a negative modulator of oVensive aggression. In addition to the ER gene knockout studies noted earlier, a recent investigation in nonhuman primates suggested several mechanisms through which ER could aVect aggression (Simon et al., 2004). More specifically, male cynomolgus macaques were fed diets containing high or low levels of soy phytoestrogens for 15 months. Those on the high soy diet exhibited significantly higher levels of agonistic behavior compared to controls (Fig. 6). Soy estrogens preferentially bind to ER, are less active than estradiolER complexes in transcriptional activity in reporter assays ( JeVerson et al., 2002), and function as ‘‘weaker agonists’’ at ER compared to naturally occurring E2 (An et al., 2001; JeVerson et al., 2002; Yi et al., 2002). These properties, combined with the alterations in agonistic behavior, suggest multiple processes for ER-driven modulation of aggression. In limbic system regions that are part of the neuroanatomical substrate for intermale aggression (Simon, 2002), a substantial portion of target neurons for estrogen express both forms of ER (Gundlah et al., 2002; Mitra et al., 2002; Shughrue and Merchenthaler, 2001). In these cells, two ER-mediated mechanisms may contribute to the modulation of agonism in males. One involves changes in the transactivation function of ER
( JeVerson et al., 2002). In keeping with this concept, studies of cell proliferation in immature mouse uterus and the regulation of cyclin D1 gene expression have shown that E2–ER complexes negatively modulate ER
induced eVects (Liu et al., 2002; Weihua et al., 2000). Another more speculative mechanism involves changes in the function of ER
/ER heterodimers. In vitro studies have shown the formation of ER
/ER heterodimers that retain DNA binding ability

FIG. 6. Frequencies (mean þ SEM) of episodes of intense aggression and submission among male cynomolgus monkeys who were on an isoflavone-free casein and lactalbumin-based diet (C/L) (n ¼ 14), a diet based in soy protein isolate containing 0.94 mg/g of isoflavone (Lo Iso) (n ¼ 15), and a diet based in soy protein isolate containing 1.88 mg/g isoflavone (Hi Iso) (n ¼ 15). * ¼ p < 0.05 relative to C/L group. From Simon et al. (2004) and reprinted with permission.

ANDROGENS AND AGGRESSION

113

(Petterson et al., 1997). The ER component of the heterodimer may normally diminish ER
-mediated transcription. In both cases, the inhibitory eVect of ER in regions, such as the medial preoptic area, bed nucleus of the stria terminalis, and medial amygdala would diminish the facilitative eVort ER
function and, as a consequence, decrease agonistic behavior. A third potential mechanism for contributions of ER to agonistic behavior involves eVects in target cells that express only this form of the receptor. The modulation of serotonergic tone in the rhesus monkey provides an example of this hypothesized process. In nonhuman primates, only ER are found in 5-HT neurons (Bethea et al., 2002; Gundlah et al., 2002). Estradiol normally acts in these cells to enhance serotonergic tone by increasing tryptophan hydroxylase synthesis and decreasing 5-HT transporter expression (Bethea et al., 2002; Lu et al., 2003). Thus, ER , as a modulator of serotonin, would decrease the propensity for aggression by maintaining normal serotonergic tone. Enhanced or reduced ER function in this region potentially could exert dramatic eVects on agonistic behavior.

IV. Hormonal Modulation of Serotonin Function

Pharmacological and molecular biological studies indicates that serotonin (5-HT) is a critical regulatory signal in the control of aggression in numerous species (Birger et al., 2003; Ferris, 2000; Kravitz, 2000; Olivier, 2004; Panksepp et al., 2003). The studies have shown that lower serotonergic tone is associated with increased aggression while enhanced serotonergic function reduces the expression of aggressive behavior. These relationships have been demonstrated in species ranging from crustaceans to rodents to primates, including humans (Birger et al., 2003; Kravitz, 2000) and the breadth of the findings has engendered a compelling basis for the extensive analysis of serotonergic tone and aggressive behavior. Broadly, agonists with selective aYnity for 5-HT1 receptors, particularly the 5-HT1A and 5-HT1B subtypes, specifically and selectively reduced oVensive intermale aggression (Olivier, 2004). From the perspective of the neuromodulator hypothesis, testicular hormones may influence behavioral activation by altering serotonin function in brain regions that either constitute or project to the neuroanatomical substrate for intermale aggression. Support for this concept can be found in both autoradiographic and in situ hybridization findings that, in combination, show overlapping distributions of estrogen-, androgen-, and serotonin-concentrating neurons as well as receptor gene expression in these regions (Herbison, 1995, 1998; Mengod et al., 1996; Simerly et al., 1990; Wright et al., 1995). Such findings, although

114

SIMON et al.

clearly of interest and suggesting interactions, are insuYcient for the establishment of neuromodulation. To test whether 5-HT function is aVected by gonad requires evidence which shows that: (1) androgens or estrogens diVerentially aVect the ability of 5-HT1A, 5-HT1B, or combined agonist treatments to alter the display of oVensive intermale aggression; (2) neuronal populations where these eVects are produced are identified; and (3) androgen or estrogen influences 5-HT1A or 5-HT1B function in these regions by altering particular aspects of serotonin function. Our laboratory conducted two investigations that assessed androgenic and estrogenic eVects on 5-HT1A and 5-HT1B functions in the context of oVensive aggression (Cologer-CliVord et al., 1997, 1999). In the initial study, systemic treatments were used to identify the relationship between functional hormonal pathways and the modulation of serotonergic eVects. Interestingly, serotonergic 1A and 1B agonists were far more eVective in reducing the display of fighting behavior in the presence of specifically acting androgens compared to estrogen. If estrogens were present, either alone or as a metabolic product, the ability of 5-HT1A and 5-HT1B agonists to inhibit oVensive aggression was restricted. When aggression was promoted by a direct androgenic eVect, however, 5-HT1A and 5-HT1B agonists were very eVective in decreasing the expression of oVensive behaviors. Neuroanatomical localization of the modulatory eVects of androgen and estrogen was assessed in a second study. Likely sites included the LS, MPOA, MAMYG, and DR based on receptor distribution maps and our understanding of neuroanatomical substrates for intermale aggression. When selective 5-HT1A and 5-HT1B agonists were microinjected into these regions, there were pronounced diVerences in the observed eVects. In the presence of diethylstilbestrol (DES), a potent specifically acting estrogen, microinjections of either 1A or 1B agonists into the LS had essentially no eVect on behavioral expression. When gonadectomized males were implanted with DHT, aggressive behavior was decreased with 1B agonist microinjection alone or in combination with the 1A agonist 8-OH-DPAT. The eVects of CGS12066B microinjection were specific because motor behavior was unaVected. At the level of the LS, then, an androgen-sensitive pathway that facilitates aggression can be attenuated by the action of serotonin at 1B receptor sites. In the MPOA, observed eVects were robust. Significantly reduced oVensive aggression in the presence of either androgen or estrogen was seen with both 5-HT1A or 5-HT1B agonist microinjections and the eVects were obtained without any impact on activity level. The MPOA may thus be a major integrative site for gonadal hormone-serotonin interactions in the regulation of T-dependent aggression. The alteration by gonadal hormones of the ability of serotonergic 1A and 1B agents to eVect T-dependent intermale aggression supports the

115

ANDROGENS AND AGGRESSION

neruomodulator hypothesis. Additional examples of comparable hormone-neurotransmitter interactions have been described in other systems, particularly in regard to reproductive behavior (Etgen, 2002; Fink et al., 1999; Melton, 2000; Uphouse, 2000), anxiety, and mood disorders (Bethea et al., 2002; Fink et al., 1998, 1999). An interesting aspect of our studies was diVerences in the ability of estrogens and androgens to attenuate the aggression-inhibiting eVects of 5-HT1A and 5-HT1B agonists and the regional variation in these eVects. If additional studies continue to find regional and cellular variation as well, regulatory models would necessarily take on an even more complex structure. Areas of inquiry requiring attention in light of our findings include mechanistic studies that address steroidal enhancement or repression of the ability of 5-HT1A and 5-HT1B agonists to attenuate oVensive intermale aggression. Estrogens can, for example, alter 5-HT1A gene expression or influence ligand availability through eVects on synthetic or degradative processes (Gundlah et al., 2002; Lu et al., 2003; McQueen et al., 1999; Mize and Alper, 2002). One step in establishing a direct eVect on 5-HT1A gene function would be the identification of a functional ERE in the promoter region of the 5-HT1A receptor gene. Interestingly, both mouse and human 5-HT1A receptor genes contain a putative ERE (Table I). The spacer element is a clear diVerence between the postulated motifs and the consensus sequence, which is five nucleotides rather than three. However, nonconsensus EREs with diVerent spacer lengths are responsive to estrogenic regulation (Berry et al., 1989; Hall et al., 2002; Klungland et al., 1994; Shupnik and Rosenzweig, 1990; Sohrabji et al., 1995). For example, the salmon GnRH and rat BDNF genes have ERE motifs with eight or nine nucleotide spacers and can bind activated estrogen receptors in vitro (Klungland et al., 1994; Sohrabji et al., 1995). TABLE I ESTROGEN RESPONSE ELEMENTS WITH VARIABLE SPACERS Species and gene Traditional spacer (n ¼ 3) Xenopus vitellogenin A2 Chicken vitellogenin II Chicken ovalbumin Human c-fos Rat prolactin Nontraditional spacer (n > 3) Rat LH- Rat BDNF Salmon GnRH Salmon GnRH Putative human 5-HT1A motif Putative mouse 5-HT1A motif

Starting position

DNA sequence


331
625
177
1209
1572

GGTCACAGTGACC GGTCAGCGTGACC GGTAACAATGTGT CGGCAGCGTGACC TGTCACTATGTCC


1173
1045
1501
1569
429
426

GGACA[N]5TGTCC GGTGA[N]9TGACC GGTCA[N]8TGTCC AGTCA[N]9TGACC GGTCA[N]5TGACC

116

SIMON et al.

Estrogen also can aVect serotonin function through processes that require multiple steps. For example, E2 treatment reportedly alters 5-HT1A receptor binding and ligand availability (Osterlund et al., 2000) as a result of modulating tryptophan hydroxylase activity and/or transporter gene expression (Bethea et al., 2000; McQueen et al., 1999; Pecins-Thompson and Bethea, 1998). It is important to recognize that to the extent that such eVects have been observed in whole animal models, they have been defined only in females. Moving from these studies to intermale aggression requires caution. In addition, ER potentially may have direct eVects on serotonergic function in DR (Alves et al., 2000; Bethea et al., 2002) and may modulate the regulatory actions of ER
.

V. Conclusions

Characterization of the hormonal processes involved in the expression of conspecific aggression has progressed diVerently in males and females. In adult males, our understanding is far more developed compared to female-typical behavior. In males, the importance of hormone metabolism has been demonstrated; aromatization and 5
-reduction of T in males are critical steps. In females, the contribution of DHEA and its metabolites as androgens may represent important system components but this has not been established experimentally. Several target neurochemical systems have been identified in males, but extensive work is needed to define the cellular processes that are aVected and the genomic and nongenomic mechanisms that mediate these eVects. A systems model of oVensive aggression that encompasses gene regulation, functional circuitry, behavioral expression, and adaptation depends on progress in this area. The neuromodulator hypothesis represents a working model that can help define the hormonal contribution to sex-typical oVensive aggression. The emphasis on neuromodulation provides a broad conceptual framework; strength is that it can incorporate findings that show that key hormonal systems produce diVerent eVects in each sex. More specifically, in males the gonadal steroids are neutral or facilitative, while in females hormonal influences seem to be largely inhibitory. Neuronanatomical substrates for aggression are still not completely defined in males, while in females little if any work has been done and characterization of cell/molecular mechanisms is in its infancy. In males, elucidating the cell/ molecular interactions among T, its metabolites, and components of the 5-HT (and no doubt other) systems is needed. In females, the various levels and processes involved in the modulation of GABAA receptor functions have not been systematically defined. These are formidable tasks, but at the same time represent only a partial list based on the representative systems covered in this chapter.

ANDROGENS AND AGGRESSION

117

Finally, we chose to focus on gonadal steroids in males and one neurosteroid in females rather than take an approach that briefly touched on the many neurotransmitters that are involved in aggression that are modulated by steroids. Other areas of interest, for example, are the eVects of corticosteroids and interactions of these and other hormones with the serotonin and vasopression systems (Ferris, 2000; Haller et al., 2000a,b). Research with animal models demonstrates the complex nature of hormonal modulation and the need for increasingly refined regulatory models of oVensive aggression. The complexity and extent of interactions also indicates that focusing on a single genetic or physiological marker as a cause of aggression is a diYcult proposition with limited utility. A systems perspective is required, one that recognizes when hormones may have a role, that physiological eVects of hormones are modulatory, and that social structure, life events, and subsequent adaptations are reflected in alterations in cellular signaling pathways. Acknowledgments

The preparation of this chapter was supported in part by grants from NIH (R01 MH59300) and the HF Guggenhiem Foundation to NGS.

References

Adamec, R. E. (1997). Transmitter systems involved in neural plasticity underlying increased anxiety and defense—implications for understanding anxiety following traumatic stress. Neurosci. Biobehav. Rev. 21, 755–765. Albert, D. J., Walsh, M. L., and Jonik, R. H. (1993). Aggression in humans: What is its biological foundation? Neurosci. Biobehav. Rev. 17, 405–425. Alves, S. E., McEwen, B. S., Hayashi, S., Korach, K. S., PfaV, D. W., and Ogawa, S. (2000). Estrogen-regulated progestin receptors are found in the midbrain raphe but not hippocampus of estrogen receptor alpha (ERalpha) gene-disrupted mice. J. Comp. Neurol. 427, 185–195. An, J., Tzagarakis-Foster, C., Scharschmidt, T. C., Lomri, N., and Leitman, D. C. (2001). Estrogen receptor -selective transcriptional activity and recruitment of coregulators by phytoestrogens. J. Biol. Chem. 276, 17808–17814. Archer, J. (1988). ‘‘The Behavioral Biology of Aggression.’’ Cambridge University Press, Cambridge, UK. Archer, J. (1991). The influence of testosterone on human aggression. Br. J. Psychol. 82, 1–28. Balthazart, J., Baeillien, M., Charlier, T. D., Cornil, C. A., and Ball, G. F. (2003). The neruoendocrinology of reproductive behavior in Japanese quail. Domest. Anim. Endrocin. 25, 69–82. Baulieu, E. E., Robel, P., and Schumacher, M. (2001). Neurosteroids: Beginning of the story. Int. Rev. Neurobiol. 46, 1–32. Berry, M., Nunez, A. M., and Chambon, P. (1989). Estrogen responsive element of the human pS2 gene is an imperfectly palindromic sequence. Proc. Natl. Acad. Sci. 86, 1218–1222.

118

SIMON et al.

Bergeron, R., de Montigny, C., and Debonnel, G. (1996). Potentiation of neuronal NMDA response induced by dehydroepiandrosterone and its suppression by progesterone: EVects mediated via sigma receptors. J. Neurosci. 16(3), 1193–1202. Bethea, C. L., Lu, N. Z., Gundlah, C., and Streicher, J. M. (2002). Diverse actions of ovarian steroids in the serotonin neural system. Front. Neuroendocrin. 23, 41–100. Bethea, C. L., Mirkes, S. J., Shively, C. A., and Adams, M. R. (2000). Steroid regulation of tryptophan hydroxylase protein in the dorsal raphe of macaques. Biol. Psychiatry 47, 562–576. Bimonte-Nelson, H. A., Singleton, R. S., Nelson, M. E., Eckman, C. B., Barber, J., Scott, T. Y., and Granholm, A. C. (2003). Testosterone, but not nonaromatizable dihydrotestosterone, improves working memory and alters nerve growth factor levels in aged male rats. Exp. Neurol. 181, 301–312. Birger, M., Swartz, M., Cohen, D., Alesh, Y., Grishpan, C., and Kotelr, M. (2003). Aggression: The testosterone-serotonin link. Isr. Med. Assoc. J. 5, 653–658. Blanchard, D. C., McKittrick, C. R., Hardy, M. P., and Blanchard, R. (2002). EVects of social stress on hormones, brain, and behavior. In ‘‘Hormones, Brain and Behavior’’ (D. PfaV, A. Arnold, A. Etgen, S. Fahrbach, and R. Rubin, Eds.), pp. 735–772. Academic Press, San Diego. Brain, P., and Haug, M. (1992). Hormonal and neurochemical correlates of various forms of animal aggression. Psychoneuroendocrinology 17, 537–551. Cologer-CliVord, A., Simon, N., and Smoluk, S. (1997). Serotonin agonist-induced decreases in intermale aggression are dependent on brain region and receptor subtype. Pharmacol. Biochem. Behav. 58, 425–430. Cologer-CliVord, A., Smoluk, S., Lu, S., and Simon, N. G. (1999). Androgens and estrogens modulate 5HT1A and 5HT1B agonist eVects on aggression. Physiol. Behav. 65, 823–828. Compagnone, N. A., and Mellon, S. H. (2000). Neurosteroids: Biosynthesis and function of these novel neuromodulators. Front. Neuroendocrinol. 21, 1–56. Choong, C. S., Kemppainen, J. A., and Wilson, E. M. (1998). Evolution of the primate androgen receptor: A structural basis for disease. J. Mol. Evol. 47, 334–342. Cui, H., Lin, S. Y., and Belsham, D. D. (2003). Evidence that deyhydroepiandrosterone, DHEA, directly inhibits GnRH gene expression in GT1-7 hypothalamic neurons. Mol. Cell. Endocrinol. 203, 13–23. Dubrovsky, B. O. (2005). Steroids, neuroactive steroids and neruosteroids in psychopathology. Prog. Neuropsychopharm. Biol. Psychiat. 29, 169–192. Elofsson, U. O., Mayer, I., Damsgard, B., and Winberg, S. (2000). Intermale competition in sexually mature arctic charr: EVects on brain monoamines, endocrine stress responses, sex hormone levels, and behavior. Gen. Comp. Endocrinol. 118, 450–460. Etgen, A. (2002). Estrogen regulation of neurotransmitter and growth factor signaling in the brain. In ‘‘Hormones, Brain and Behavior’’ (D. PfaV, A. Arnold, A. Etgen, S. Fahrbach, and R. Rubin, Eds.), pp. 381–440. Academic Press, San Diego. Ferris, C. F. (2000). Adolescent stress and neural plasticity in hamsters: A vasopressin-serotonin model of inappropriate aggressive behaviour. Exp. Physiol. 85, S85–S90. Fink, G., Sumner, B. E., McQueen, J. K., Wilson, H., and Rosie, R. (1998). Sex steroid control of mood, mental state and memory. Clin. Exp. Pharmacol. Physiol. 25, 764–775. Fink, G., Sumner, B. E., Rosie, R., Wilson, H., and McQueen, J. K. (1999). Androgen actions on central serotonin neurotransmission: Relevance for mood, mental state and memory. Behav. Brain Res. 105, 53–68. Freeman, E. R., Bloom, D. A., and McGuire, E. J. (2001). A brief history of testosterone. J. Urol. 165, 371–373. Fusani, L., Metzdorf, R., Hutchison, J. B., and Gahr, M. (2003). Aromatase inhibition aVects testosterone-induced masculinization of song and the neural song sytem in female canaries. J. Neurobiol. 54, 370–379.

ANDROGENS AND AGGRESSION

119

Giammanco, M., Tabacchi, G., Giammano, S., Di Majo, D., and La Guardia, M. (2005). Testosterone and aggressiveness. Med. Sci. Monit. 11, 136–145. Godwin, J., and Crews, D. (2002). Hormones, brain, and behavior in reptiles. In ‘‘Hormones, Brain and Behavior’’ (D. PfaV, A. Arnold, A. Etgen, S. Fahrbach, and R. Rubin, Eds.), pp. 545–586. Academic Press, San Diego. Gould, E., and Cameron, H. A. (1997). Early NMDA receptor blockade impairs defensive behavior and increases cell proliferation in the dentate gyrus of developing rats. Behav. Neurosci. 111, 49–56. Gundlah, C., Lu, N. Z., and Bethea, C. L. (2002). Ovarian steroid regulation of monoamine oxidaseA and -B mRNAs in the macaque dorsal raphe and hypothalamic nuclei. Psychopharmacology 160, 271–282. Hall, J. M., McDonnell, D. P., and Korach, K. S. (2002). Allosteric regulation of estrogen receptor structure, function, and coactivator recruitment by diVerent estrogen response elements. Mol. Endocrinol. 16, 469–486. Haller, J., Halasz, J., Mikics, E., Kruk, M. R., and Makara, G. (2000a). Ultradian corticosterone rhythm and the propensity to behave aggressively in male rats. J. Neuroendocrinol. 12, 937–940. Haller, J., Millar, S., van de Schraaf, K., de Kloet, R. E., and Kruk, M. R. (2000b). The active phaserelated increase in corticosterone and aggression are linked. J. Neuroendocrinol. 12, 431–436. Hau, M., Stoddard, S. T., and Soma, K. K. (2004). Territorial aggression and hormones during the non-breeding season in a tropical bird. Horm. Behav. 45, 40–49. Herbison, A. E. (1995). Sexually dimorphic expression of androgen receptor immunoreactivity by somatostatin neurons in rat hypothalamic periventricular nucleus and bed nucleus of the stria terminalis. J. Neuroendrocrinol. 7, 543–553. Herbison, A., and Fenelon, V. (1995). Estrogen regulation of GABA-A receptor subunit mRNA expression in preoptic area and bed nucleus of the stria terminalis of female rat brain. J. Neurosci. 15, 2328–2337. Hutton, L., Guibao, G., and Simerly, R. (1998). Development of a sexually dimorphic projection form the bed nuclei of the stria teminalis to the anteroventral periventricular nucleus in the rat. J. Neurosci. 18, 3003–3013. Imamura, M., and Prasad, C. (1998). Modulation of GABA-gated chloride ion influx in the brain by dehydroepiandrosterone and its metabolites. Biochem. Biophys. Res. Commun. 243(3), 771–775. JeVerson, W. N., Padilla-Banks, E., Clark, G., and Newbold, R. R. (2002). Assessing estrogenic activity of phytochemicals using transcriptional activation and immature mouse uterotrophic responses. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 777, 179–189. Jellinck, P. H., Lee, S. J., and McEwen, B. S. (2001). Metabolism of dehydroepiandrosterone by rat hippocampal cells in culture: Possible role of aromatization and 7-hydroxylation in neuroprotection. J. Steroid Biochem. Mol. Biol. 78, 313–317. Jellinck, P. H., Crof, G., McEwen, B. S., Gottfried-Blackmore, A., Jones, G., Byford, V., and Bulloch, K. (2005). Metabolism of dehydroepiandrosterone by rodent brain cell lines: Relationship between 7-hydroxylation and aromatization. J. Steroid Biochem. Mol. Biol. 93, 81–86. Katzenellenbogen, B. S. (2000). Mechanism of action and cross-talk between estrogen receptor and progesterone receptor pathways. J. Soc. Gynecol. Investing. 7, S33–S37. Klungland, H., Anderson, O., Kisen, G., Alestrom, P., and Tora, L. (1994). Estrogen receptor binds to the salmon GnRH gene in a region with long palindromic sequences. Mol. Cell. Endocrinol. 95, 147–154. Koehler, K. F., Helguero, L. A., Haldosen, L. A., Warner, M., and Gustafsson, J. A. (2005). Reflections on the discovery and significance of estrogen receptor beta. Endocr. Rev. 26, 465–478. Kravitz, E. A. (2000). Serotonin and aggression: Insights gained from a lobster model system and speculations on the role of amine neurons in a complex behavior. J. Comp. Physiol. 186, 221–238.

120

SIMON et al.

Kuiper, G. G. J. M., Carlson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S., and Gustafsson, J.-A. (1997). Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors
and . Endocrinology 138, 863–870. Labrie, F. (2003). Extragonadal synthesis of sex steroids: Intracrinology. Ann. Endocrinol. 64, 95–107. Lee, H. J., and Chang, C. (2003). Recent advances in androgen action. Cell Mol. Life Sci. 60, 1613–1622. Liu, S., Sugimoto, Y., Kulp, S. K., Jiang, J., Chang, H. L., Park, K. Y., Kashida, Y., and Lin, Y. C. (2002). Estrogenic down-regulation of protein tyrosine phosphatase gamma (PTP gamma) in human breast is associated with estrogen receptor alpha. Anticancer Res. 22, 3917–3923. Lu, N. Z., Eshleman, A. J., Janowsky, A., and Bethea, C. L. (2003). Ovarian steroid regulation of serotonin reuptake transporter (SERT) binding, distribution, and function in female macaques. Mol. Psychiatry 8, 353–360. Lu, S., Kaplna, J. R., Adams, M. R., and Simon, N. G. (2006). Hormonal regulation of neural androgen receptor in cynomolgus monkey. J. Neurobiol., under review. Lu, S., Mo, Q., Hu, S., Garippa, C., and Simon, N. G. (2003). Dehydroepiandrosterone upregulates neural androgen receptor and transcriptional activity. J. Neurobiol. 57, 163–171. Lu, S., Simon, N., McKenna, S., Nau, G., and Cologer-CliVord, A. (1998). Neural androgen receptor: Sex diVerences and similarities in autoregulation. Endocrinology 139, 1594–1601. Majewska, M. D. (1992). Neurosteroids: Endogenous bimodal modulators of the GABAA receptor. Mechanism of action and physiological significance. Prog. Neurobiol. 38(4), 379–395. Majewska, M. D. (1995). Neuronal actions of dehydroepiandrosterone: Possible roles in brain development, aging, memory, and aVect. Ann. N. Y. Acad. Sci. 774, 111–120. Majewska, M. D., and Schwartz, R. D. (1987). Pregnenolone sulfate: An endogenous antagonist of the gamma-aminobutyric acid receptor complex in brain? Brain Res. 404, 355–360. Majewska, M. D., Demirgoren, S., and London, E. D. (1990). Binding of pregnenolone sulfate to rat brain membranes suggests multiple sites of steroid action at the GABAA receptor. Eur. J. Pharmacol. 189, 307–315. Matthews, J., and Gustafasson, J. A. (2003). Estrogen signaling: A sublte balnce between ER alpha and ER beta. Mol. Interv. 5, 281–292. Maurice, T., Phan, V. L., Urani, A., Kamei, H., Noda, Y., and Nabeshima, T. (1999). Neuroactive neurosteroids as endogenous eVectors for the sigma1 (sigma1) receptor: Pharmacological evidence and therapeutic opportunities. Jpn. J. Pharmacol. 81(2), 125–155. McQueen, J. K., Wilson, H., Sumner, B. E. H., and Fink, G. (1999). Serotonin transporter (SERT) mRNA and binding site densities in male rat brain aVected by sex steroids. Brain Res. Mol. Brain Res. 63, 241–247. Mehta, A. K., and Ticku, M. K. (1999). An update on GABA-A receptors. Brain Res. 29, 196–217. Melcangi, R. C., Poletti, A., Cavarretta, I., Celotti, F., Colciago, F., Magnaghi, V., Motta, M., NegriCesi, P., and Martini, L. (1998). The 5alpha-reductase in the central nervous system: Expression and modes of control. J. Steroid Biochem. Mol. Biol. 65, 295–299. Melton, L. (2000). Sex is all in the brain: Report of a Novartis foundation symposium on the neuronal and cognitive eVects of oestrogens, London, UK. Trends Endocrinol. Metab. 11, 69–71 (September 7–9, 1999). Mengod, G., Vilaro, T., Raurich, A., Luopez-Gimuenez, F., Cortues, R., and Palacios, J. (1996). 5-HT receptors in mammalian brain: Receptor autoradiograpphy and in situ hybridization studies of new ligands and newly identified receptors. Histochem. J. 11, 747–758. Miczek, K. A., Fish, E. W., and De Bold, J. F. (2003). Neurosteroids, GABA-A receptors, and escalated aggressive behavior. Horm. Behav. 44, 242–257. Mitra, S. W., Hoskin, E., Yudkovitz, J., Pear, L., Wilkinson, H. A., Hayashi, S., PfaV, D. W., Ogawa, S., Rohrer, S. P., SchaeVer, J. M., McEwen, B. S., and Alves, S. E. (2002). Immunolocalization of

ANDROGENS AND AGGRESSION

121

estrogen receptor h in the mouse brain: Comparison with estrogen receptor. Endocrinology 144, 2055–2067. Mize, A. L., and Alper, R. H. (2002). Rapid uncoupling of serotonin-1A receptors in rat hippocampus by 17beta-estradiol in vitro requires protein kinases A and C. Neuroendocrinology 76, 339–347. Mo, Q., Lu, S., Hu, S., and Simon, N. G. (2004). DHEA and DHEA sulfate diVerentially regulate androgen receptor and its transcriptional activity. Mol. Brain Res. 126, 165–172. Naftolin, F., Horvath, T. L., and Balhazart, J. (2001). Estrogen synthetase (aromatase) immunohistochemistry reveals concordance between avian and rodent limbic systems and hypothalami. Exp. Biol. Med. 226, 717–725. Nelson, R. J. (1995). ‘‘An Introduction to Behavioral Endocrinology.’’ Sinaur Assoc., Sunderland, MA. Ogawa, S., Inoue, S., Watanabe, T., Hiroi, H., Orimo, A., Hosoi, T., Ouchi, Y., and Muramatsu, M. (1998). The complete primary structure of human estrogen receptor beta (hER beta) and its heterodimerization with ER alpha in vivo and in vitro. Biochem. Biophys. Res. Comm. 243, 122–126. Olivier, B. (2004). Serotonin and aggression. Ann. N. Y. Acad. Sci. 1036, 382–392. Osterlund, M. K., Halldin, C., and Hurd, Y. L. (2000). EVects of chronic 17beta-estradiol treatment on the serotonin 5-HT(1A) receptor mRNA and binding levels in the rat brain. Synapse 35, 39–44. Panksepp, J. B., Yue, Z., Drerup, C., and Huber, R. (2003). Amine neurochemistry and aggression in crayfish. Microsc. Res. Tech. 60, 360–368. Pecins-Thompson, M., and Bethea, C. L. (1998). Ovarian steroid regulation of 5-HT1A autoreceptor messenger ribonucleic acid expression in the dorsal raphe of rhesus macaques. Neuroscience 89, 267–277. Petterson, K., Grandien, K., Kuiper, G. G. J. M., and Gustafsson, J. (1997). Mouse estrogen receptor h forms estrogen response element-binding heterodimers with estrogen receptor a. Mol. Endocrinol. 11, 1486–1496. PfaV, D., Ogawa, S., Kia, K., Vasudevan, N., Krebs, C., Frohlich, J., and Kow, L. (2002). Genetic mechanisms in neural and hormonal controls over female reproductive behavior. In ‘‘Hormones, the Brain, and Behavior’’ (D. PfaV, A. Arnold, A. Etgen, B. Fahrbach, and R. Rubin, Eds.), pp. 441–510. Academic Press, San Diego. Rasika, S., Alveraz-Buylla, A., and Nottebohm, F. (1999). BDNF mediates the eVects of testosterone on the survival of new neurons in an adult brain. Neuron 22, 53–62. Robel, P., and Baulieu, E. E. (1995). Dehydroepiandrosterone (DHEA) is a neuroactive neurosteroid. Ann. N. Y. Acad. Sci. 774, 82–110. Rupprecht, R. (2003). Neuroactive steroids: Mechanisms of action and neuropsychopharmacological properties. Psychoneuroendocrinology 28, 139–168. Rupprecht, R., di Michele, F., Hermann, B., STrohle, A., Lancel, M., Romeo, E., and Holsboer, F. (2001). Neuroactive steroids: Molecular mcehanisms of action and implications for neuropsychopharmacology. Brain Res. Brain Res. Rev. 37, 59–67. Sato, T., Matsumoto, T., Kawano, H., Watanabe, T., Uematsu, Y., Sekine, K., Fukuda, T., Aihara, K., Krust, A., Yamada, T., Nakamichi, Y., Yamamoto, Y., et al. (2004). Brain masculinization requires androgen receptor function. Proc. Natl. Acad. Sci. 101, 1673–1678. Segovia, S., and Guillamon, A. (1993). Sexual dimorphism in the vomeronasal pathway and sex diVerences in reproductive behaviors. Brain Res. Brain Res. Rev. 18, 51–74. Shughrue, P. J., Lane, M. V., and Merchenthaler, I. (1997). Comparative distribution of estrogen receptor-alpha and -beta mRNA in the rat central nervous system. J. Comp. Neurol. 388, 507–525. Shughrue, P. J., Lane, M. V., Scrimo, P. J., and Merchenthaler, I. (1998). Comparative distribution of estrogen receptor-
(ER-
) and  (ER-) mRNA in the rat pituitary, gonad and reproductive tract. Steroids 63, 498–504.

122

SIMON et al.

Shughrue, P. J., and Merchenthaler, I. (2001). Distribution of estrogen receptor beta immunoreactivity in the rat central nervous system. J. Comp. Neurol. 436, 64–81. Shupnik, M. A., and Rosenzweig, B. J. (1990). Identification of an estrogen-responsive element in the rat LH beta gene: DNA-estrogen receptor interactions and functional analysis. Biol. Chem. 266, 17084–17091. Siegel, A., Roeling, T. A., Gregg, T. R., and Kruk, M. R. (1999). Neuropharmacology of brainstimulation-evoked aggression. Neurosci. Biobehav. Rev. 23, 359–389. Sierens, J. E., Scobie, G. A., Wilson, J., and Saunders, P. T. (2004). Cloning of oestrogen receptor beta from old and new world primates: Identification of splice variants and functional analysis. J. Mol. Endocrinol. 32, 703–718. Silverin, B., Baillien, M., Foidart, A., and Balthazart, J. (2000). Distribution of aromatase activity in the brain and peripheral tissues of passerine and nonpasserine avian species. Gen. Comp. Endocrinol. 117, 34–53. Simerly, R. B. (1998). Organization and regulation of sexually dimorphic neuroendocrine pathways. Behav. Brain Res. 92, 195–203. Simerly, R. B., Chang, C., Muramatsu, M., and Swanson, L. W. (1990). Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: An in situ hybridization study. J. Comp. Neurol. 294, 76–95. Simon, N. G. (2002). Hormonal processes in the development and expression of aggressive behavior. In ‘‘Hormones, Brain and Behavior’’ (D. PfaV, A. Arnold, A. Etgen, S. Fahrbach, and R. Rubin, Eds.), pp. 339–391. Academic Press, San Diego. Simon, N. G., Kaplan, J. R., Hu, S., Register, T. C., and Adams, M. R. (2004). Increased aggressive behavior and decreased aYliative behavior in adult male monkeys after long-term consumption of diets rich in soy protein and isoflavones. Horm. Behav. 45, 278–284. Simon, N., McKenna, S., Lu, S., and Cologer-CliVord, A. (1996). Development and expression of hormonal systems regulating aggression. Ann. N. Y. Acad. Sci. 794, 8–17. Sohrabji, F., Miranda, R. C. G., and Toran-Allerand, C. D. (1995). Identification of a putative estrogen response element in the gene encoding brain-derived neurotrophic factor. Proc. Natl. Acad. Sci. USA 92, 11110–11114. Soma, K. K., and Wingfield, J. C. (2001). Dehydroepiandrosterone in songbird plasma: Seasonal regulation and relationship to territorial aggression. Gen. Comp. Endocrinol. 123, 144–155. Sousa, A., and Ticku, M. K. (1997). Interactions of the neurosteroid dehydroepiandrosterone sulfate with the GABA(A) receptor complex reveals that it may act via the picrotoxin site. J. Pharmacol. Exp. Ther. 282(2), 827–833. Sun, J., Meyers, M., Fink, B. E., Rajendran, R., Katzenellenbogen, J. A., and Katzenellenbogen, B. S. (1999). Novel ligands that function as selective estrogens or antiestrogens for estrogen receptor-
or estrogen receptor-. Endocrinology 140, 800–804. Uphouse, L. (2000). Female gonadal hormones, serotonin, and sexual receptivity. Brain Res. Brain Res. Rev. 33, 242–257. van Goozen, S. H., van den Ban, E., Matthys, W., Cohen-Kettenis, P. T., Thijssen, J. H., and van Engeland, H. (2000). Increased adrenal androgen functioning in children with oppositional defiant disorder: A comparison with psychiatric and normal controls. J. Am. Acad. Child Adolesc. Psychiatry 39(11), 1446–1451. von Engelhardt, N., Kappeler, P. M., and Heistermann, M. (2000). Androgen levels and female social dominance in Lemur catta. Proc. Biol. Soc. 267, 1533–1539. Weihua, Z., Saji, S., Makinen, S., Cheng, G., Jensen, E. V., Warner, M., and Gustafsson, J. A. (2000). Estrogen receptor (ER) beta, a modulator of ERalpha in the uterus. Proc. Natl. Acad. Sci. USA 97, 5936–5941.

ANDROGENS AND AGGRESSION

123

Wilkinson, H. A., Dahllund, J., Liu, H., Yudkovitz, J., Cai, S. J., Nilsson, S., SchaeVer, J. M., and Mitra, S. W. (2002). Identification and characterization of a functionally distinct form of human estrogen receptor beta. Endocrinology 143, 1558–1561. Wingfield, J. C., Jacobs, J., and Hillgarth, N. (1997). Ecological constraints and the evolution of hormone-behavior interrelationships. Ann. N. Y. Acad. Sci. 807, 22–41. Woolley, S. C., Sakata, J. T., and Crews, D. (2004). Evolutionary insights into the regulation of courtship behavior in male amphibians an reptiles. Physiol. Behav. 83, 347–360. Wright, D. E., Seroogy, K. B., Lundgren, K. H., Davis, B. M., and Jennes, L. (1995). Comparative localization of serotonin lA, 1C and 2 receptor subtype mRNAs in rat brain. J. Comp. Neurol. 351, 357–373. Yang, L. Y., and Arnold, A. P. (2000). Interaction of BDNF and testosterone in the regulation of adult perineal motoneurons. J. Neurobiol. 44, 308–319. Yang, L. Y., Verhovshek, T., and Sengelaub, D. R. (2004). Brain-derived neurotrophic factor and androgen interact in the maintenance of dendritic morphology in a sexually dimorphic rat spinal nucleus. Endocrinology 145, 161–168. Yi, P., Bhagat, S., Hilf, R., Bambara, R. A., and Muyan, M. (2002). DiVerences in the abilities of estrogen receptors to integrate activation functions are critical for subtype-specific transcriptional responses. Mol. Endocrinol. 16, 1810–1827. Young, J., Corpechot, C., Perche´, F., Haug, M., Baulieu, E. E., and Robel, P. (1995). Neurosteroids: Pharmacological eVects of a 3
-hydroxy-steroid dehydrogenase inhibitor. Endocrine 2, 505–509. Young, J., Corpechot, C., Perche´, F., Eychenne, B., Haug, M., Baulieu, E. E., and Robel, P. (1996). Neurosteroids in the mouse brain: Behavioral and pharmacological eVects of a 3aˆ-hydroxysteroid dehydrogenase inhibitor. Steroids 61, 144–149.