Modifications of Testosterone-dependent Behaviors by Estrogen Receptor-␣ Gene Disruption in Male Mice*

The role of the a form of estrogen receptor (ERa) gene expression in the regulation of testosterone-dependent male reproductive behaviors was investigated using ER knockout mice (ERKO), which are specifically deficient in functional ERa, but not ERb, gene expression. Previous studies in gonadally intact ERKO mice revealed that male aggressive behavior was greatly reduced by the lack of a functional ERa gene. In the present study the almost complete suppression of male-typical offensive attacks was further confirmed in ERKO mice that had been singly housed since weaning. Regarding aggression, it was also found that ERa gene disruption virtually abolished the propensity to initiate offensive attacks, even though ERKO mice could elicit attacks from resident C57BL/6J mice as wild-type (WT) and heterozygous littermates. Daily injection of testosterone propionate (TP) was completely ineffective in inducing aggressive behavior in gonadectomized ERKO mice, whereas it successfully restored aggression in WT mice. In contrast, male sexual behaviors, mounts and intromissions, were induced by daily injection of TP in both gonadectomized ERKO and WT mice. In addition to TP, dihydrotestosterone propionate (DHTP) was also effective in restoring mounts in ERKO mice, although DHTP was much more potent in WT mice than in ERKO mice. Neither TP nor DHTP, however, ever induced ejaculation in ERKO mice. These results together with previous findings in gonadally intact ERKO mice suggest that ERa may be responsible for the regulation by testosterone of consummatory, but not motivational, aspects of male sexual behavior. Finally, ERKO male mice retrieved newborn pups placed in their home cage with similar latencies to males of the two other genotypes. During parental behavior tests, however, a higher percentage of ERKO mice (70%) showed infanticide compared with WT mice (35%). The latter result was interpreted as showing that ERa activation by testosterone during the perinatal period may exert a suppressive effect on testosteroneinducible infanticide in adulthood. With respect to three major testosterone-dependent behavioral systems reflecting masculinization, these findings demonstrate three different types of effects due to ERa gene disruption. (Endocrinology 139: 5058–5069, 1998) I IS WELL known that testosterone regulates various behavioral and neuroendocrine functions in male mice. In the brain, as in many peripheral tissues, testosterone not only acts through androgen receptors (AR) in its original form or as the 5a-reduced form (dihydrotestosterone), but also partly through estrogen receptors (ER), after being aromatized to estrogen. The relative importance of these two mechanisms in specific reproductive behaviors, e.g. sexual, aggressive, and parental, has been studied extensively during the last 3 decades. These studies included the comparisons of potency of testosterone with its 5a-reduced (i.e. dihydrotestosterone) or aromatized metabolites (i.e. estradiol), testosterone treatment in conjunction with aromatization inhibitors, or specific receptor antagonists for androgen or estrogen receptors. Recently, however, the second form of ER, ERb, was cloned (1, 2), and subsequently its messenger RNA (mRNA)/ protein was localized in various brain regions (3–5). Although its exact functions and mechanisms of action are not yet determined, it is known to bind to 17b-estradiol with a similar affinity to that of the classical ER (6, 7), now termed ERa. Therefore, our knowledge about ER-dependent actions of testosterone needs to be reevaluated. One direct way is to actively manipulate the gene expression of specific ER through the use of antisense DNA (8) or gene-targeting methods. Especially useful for the latter approach are ERa genedeficient (ERKO) mice (9, 10), which lack functional ERa, but have ERb (5, 11) genes. They provide us with a great opportunity to study the role of ERa in a number of different behaviors and in a number of different endocrine conditions. Previously, we have characterized behavioral modifications induced by the lack of functional ERa in gonadally intact male (12) and female (13) ERKO mice. These studies revealed that ERKO male mice almost completely lacked ejaculatory behavior as well as male-typical offensive attacks, whereas some components of male sexual behavior remained intact. In the present study we have examined the effects of ERa gene disruption on male reproductive behaviors by controlling their previous social experience as well as their gonadal conditions, i.e. tested after gonadectomy and androgen