Conversion of androgen to estrogen in the rat brain is catalyzed by aromatase enzymes. The maximum concentrations of these enzymes are found within the hypothalamus and amygdala, where they appear to play an important role in the process by which androgens affect both behavior and neuroendocrine function. In the present study, we measured the levels of aromatase activity (AA) in 20 nuclei and brain regions of the adult rat brain. Individual nuclei were microdissected from 600-micron frozen sections. Tissues from 3 animals were pooled, and AA was measured by an in vitro radiometric assay that quantifies the stereospecific production of 3H2O from [1 beta-3H]androstenedione as an index of estrogen formation. We report that AA is heterogeneously distributed within the rat brain. The greatest amounts of activity were found in the bed nucleus (n.) of the stria terminalis (700 protein fmol/h . mg) and in the medial (MA) and cortical amygdala (400-600 fmol/h . mg protein) of the male. There was an evident rostral-caudal and medial-lateral gradient in AA throughout the diencephalon. Activity was high in the periventricular preoptic n. and medial preoptic n.; intermediate in the suprachiasmatic preoptic n., anterior hypothalamus, periventricular anterior hypothalamus, and ventromedial n.; and low in the arcuate n.-median eminence, lateral preoptic n., supraoptic n., dorsomedial n., and lateral hypothalamus. Regions devoid of measurable AA included the medial and lateral septum, caudate-putamen, hippocampus, and parietal cortex. In the female, AA was greatest in the MA and cortical amygdala. We found that AA in the MA, stria terminalis n., suprachiasmatic preoptic n., periventricular preoptic in., medial preoptic n., anterior hypothalamus, and ventromedial n. was significantly greater (P less than 0.05) in males than in females. Orchidectomy reduced AA to levels seen in females, and administration of testosterone to castrated males restored AA in these areas. No significant sex differences were observed in any other hypothalamic or amygdaloid nuclei, although AA was increased by testosterone treatment in the periventricular anterior hypothalamus, arcuate n.-median eminence, and lateral hypothalamus. Our results provide a quantitative profile of AA in specific hypothalamic and limbic nuclei of the rat brain as well as information on the control of AA within these discrete regions.

We studied the distribution and regulation of aromatase activity in the adult rat brain with a sensitive in vitro assay that measures the amount of 3H2O formed during the conversion of [1 beta-3H]androstenedione to estrone. The rate of aromatase activity in the hypothalamus-preoptic area (HPOA) was linear with time up to 1 h, and with tissue concentrations up to 5 mgeq/200 microliters incubation mixture. The enzyme demonstrated a pH optimum of 7.4 and an apparent Michaelis-Menten constant (Km) of 0.04 microns. We found the greatest amount of aromatase activity in amygdala and HPOA from intact male rats. The hippocampus, midbrain tegmentum, cerebral cortex, cerebellum, and anterior pituitary all contained negligible enzymatic activity. Castration produced a significant decrease in aromatase activity in the HPOA (P less than 0.001), but not in the amygdala or cerebral cortex (P greater than 0.05). The HPOAs of male rats contained significantly greater aromatase activity than the HPOAs of female rats. In females, this enzyme activity did not change during the estrous cycle or after ovariectomy. Administration of testosterone to gonadectomized male and female rats significantly enhanced HPOA aromatase activities (P less than 0.05) to levels approximating those found in HPOA from intact males. Therefore, our results suggest that testosterone, or one of its metabolites, is a major steroidal regulator of HPOA aromatase activity in rats.

SUMMARY Testosterone and androstenedione levels in plasma and testicular tissue of developing rats were measured using gas—liquid chromatography with electron capture detection. The major androgen of both the adult and early postnatal period of development was testosterone. Pooled plasma from 230 one-day-old male rats contained 0·027 μg. testosterone/100 ml. The concentration of testosterone in the testes at this age was 0·328 μg./g. wet tissue. With increasing age there was a decline in testosterone concentration in plasma as well as in gonadal tissue which lasted until about the age of 30 days. The period from 40 to 60 days was characterized by an increasing concentration of testosterone in the plasma and gonads. During adulthood, testosterone reached concentrations as high as 0·202 μg./100 ml. peripheral plasma. Androstenedione could not be detected in the circulation during the critical period of neonatal neural sexual differentiation, but it was present in the testes at this stage. In the pubertal and adult stages androstenedione was found in the plasma and testes. Its concentration, particularly in adulthood, was not as great as that of testosterone. These results indicate that testosterone is present in plasma and testicular tissue of the rat during the neonatal period when behavioural and physiological sexual differentiation is presumed to occur.

Sheep are one of the few animal models in which natural variations in male sexual preferences have been studied experimentally. Approximately 8% of rams exhibit sexual preferences for male partners (male-oriented rams) in contrast to most rams, which prefer female partners (female-oriented rams). We identified a cell group within the medial preoptic area/anterior hypothalamus of age-matched adult sheep that was significantly larger in adult rams than in ewes. This cell group was labeled the ovine sexually dimorphic nucleus (oSDN). In addition to a sex difference, we found that the volume of the oSDN was two times greater in female-oriented rams than in male-oriented rams. The dense cluster of neurons that comprise the oSDN express cytochrome P450 aromatase. Aromatase mRNA levels in the oSDN were significantly greater in female-oriented rams than in ewes, whereas male-oriented rams exhibited intermediate levels of expression. Because the medial preoptic area/anterior hypothalamus is known to control the expression of male sexual behaviors, these results suggest that naturally occurring variations in sexual partner preferences may be related to differences in brain anatomy and capacity for estrogen synthesis.

We studied the regulation of aromatase activity in the hypothalamus-preoptic area ( HPOA ) of adult male rats using a sensitive in vitro assay which measures the amount of 3H2O formed by tissue homogenates during the conversion of [1 beta-3H]androstenedione to estrone. After castration, HPOA aromatase activity was decreased by 60% (P less than 0.05), seminal vesicle (SV) and ventral prostate (VP) weights were significantly decreased (P less than 0.05), and serum LH levels were elevated. We found that testosterone (T) or 5 alpha-dihydrotestosterone (DHT) administered in Silastic capsules for 7 days reversed the effects of castration. Testosterone and DHT stimulated HPOA aromatase activity 133% and 92%, respectively (P less than 0.05). Both steroids significantly increased SV and VP wet weights and suppressed serum levels of LH (P less than 0.05). Administration of either estradiol or progesterone did not reverse the effect of castration on HPOA aromatase activity or any other parameter measured. To determine the involvement of androgen receptors in the mechanism by which androgens affect brain aromatase, we administered the nonsteroidal antiandrogen flutamide to intact male rats (15 mg/day for 7 days). There was 42% less HPOA aromatase activity in treated rats than in oil-injected controls (P less than 0.05). Flutamide significantly decreased SV and VP wet weights, while serum LH levels were enhanced (P less than 0.05). Likewise, administration of flutamide to T-implanted castrated males blocked the T-induced increase in HPOA aromatase activity and accessory sexual organ wet weights, and prevented the T-induced suppression of serum LH. Flutamide given alone to castrated rats had no effect. Since both T and DHT stimulated HPOA aromatase activity and since the effects of T are blocked by the concomitant administration of the antiandrogen flutamide, we concluded that the control of HPOA aromatase activity by androgens is receptor mediated.