AbstractIn the rat, there is a marked but transient increase in hypothalamic aromatase activity during the last week of fetal life. The present study was undertaken to gain insight into the regulation of this developmental pattern. Hypothalamic fragments comprising the medial basal hypothalamus and the suprachiasmatic region (henceforth referred to as preoptic area) were explanted and cultured in serum‐free medium for 2 to 5 days. Aromatase activity was measured by the formation of3H2O, utilizing either [1ß‐3H]androstene‐dione or [1ß‐3H]testosterone as substrate. Maximal rates of activity were obtained at a saturating concentration of 0.3 μM [1ß‐3H]testosterone. Confirmation of the identity of the [3H]estradiol formed was demonstrated by recrystallization of the derivatized estradiol to constant specific activity following incubation with [1,2,6,7‐3H]testosterone. In agreement with previous reports,in vivohypothalamic aromatase activity was negligible before gestational day (GD) 16, increased strikingly by GD19 (>5.0 pmol/h/mg protein) and decreased, thereafter, to low levels at GD22 (∼1.0 pmol/h/mg protein). Medial basal hypothalamus‐preoptic area fragments explanted before GD17 failed to develop aromatase activityin vitro. If the tissue was explanted on GD17 or 18 (i.e. when the in vivo rate of activity was increasing), the enzyme activity did not continue to increase, but it was rather maintained for 2 days before decreasing in a manner that closely mimicked the decline observed in vivo. A similar, butimmediate decline was observed when the tissue was explanted on GD19 (i.e. at the time when theactivity peaksin vivo). Exposure of explants to either growth factors (insulin‐like growth factor II, epidermal growth factor, and basic or acidic fibroblast growth factor), or steroids (estradiol‐17ß, progesterone, testosterone, dihydrotestosterone and corticosterone) failed to either increase aromatase activity before the peak at GD19 or ameliorate its perinatal decline. Increase of Ca2+fluxes with the ionophore A23187 or activation of the cyclic AMP, cyclic GMP, or protein kinase C pathways were similarly ineffective, as was angiotensin II, a recently proposed stimulator of neural aromatase. In contrast, aromatase activity was suppressed 2‐ to 4‐fold by activation of the cyclic AMP pathway (with either forskolin or 8‐bromo‐cyclic AMP) or by the androgens, testosterone and dihydrotestosterone.These results suggest that: 1) the appearance of aromatase activity in the rat hypothalamus before GD17 requires the unfolding of extrahypothalamic events, 2) the increase in aromatase activity that occurs before GD19 also requires extrahypothalamic inputs and does not involve any of the known intracellular signal transduction pathways, and 3) the decline in activity observed after GD19 is regulated within the hypothalamus, and appears to be determined, at least in part, by the activation of cyclic AMP formation. A potential