It is clear that much of the masculinization of the brain in rats and mice is mediated by aromatized metabolites of testicular androgens acting upon estrogen receptors (ERs). For example, exogenous estrogens, which presumably exert little effect on androgen receptors (ARs), can reverse the loss of masculine behavior and neural morphology in males that have been castrated, both in development and adulthood. However, we find that rats and mice carrying a dysfunctional AR gene, so-called testicular feminization mutation (Tfm) males, are partly or completely demasculinized in terms of at least one non-reproductive behavior and each of the numerous brain regions we have examined so far. These findings indicate that in fact AR normally plays a role in the masculinization of at least some behaviors, and potentially every brain region, in rodents.
For example, the medial amygdala (MeA) is about 150% larger in volume in wildtype (wt) male rats than in wt females. Tfm males display an intermediate volume, significantly greater than wt females yet significantly less than wt males. Astrocytes in the posterodorsal portion of the MeA (MePD) of rats are also sexually dimorphic, both in number and arbor complexity, and Tfm males are wholly feminine in these features. Likewise, in our measurements of sexually dimorphic characters in the ventromedial hypothalamus (VMH), the suprachiasmatic nucleus (SCN), and the paraventricular nucleus (PVN), Tfm males are wholly feminine. Even in the sexually dimorphic nucleus of the preoptic area (SDNPOA), where the volume is masculine in Tfm males, the size of the neurons is nevertheless reduced in Tfm males compared to wt males.
It is difficult to assess masculine reproductive behavior in Tfm males because they have an entirely feminine exterior phenotype, with a clitoris, vagina, etc. Nevertheless, they have been reported to show many masculine reproductive behaviors, as would be expected if those were mediated by ERs. However, we find that anxiety-related behaviors, such as measured in an open field with a novel object, the elevated plus maze, and the light/dark box, are greater in Tfm males than in wt males in both rats and mice. Tfm animals also show a heightened corticosterone response to mild stress. These results suggest that masculinization of anxiety-related behavior is heavily reliant on stimulation of AR, presumably in the brain. We are exploring the sites of AR action by use of Cre- lox technology to delete AR in selective tissues.
We are using the same technology to explore the site(s) of androgen action on the spinal nucleus of the bulbocavernosus (SNB), a group of motoneurons that innervate two striated muscles, the bulbocavernosus and levator ani (BC/LA), which are attached to the base of the penis. By selectively deleting AR in either motoneurons alone, or in muscle fibers alone, we hope to understand how androgen spares this system from apoptosis in development, and regulates neural plasticity of the motoneurons in adulthood.
