Although glucocorticoid hormones have important roles in the development of neurotransmitter systems in cells derived from the neural crest, it is not known whether they have parallel effects on neuronal development in the brain. To address this issue, we have established an in vitro system of fetal medulla oblongata (MO) to follow development of the epinephrine-synthesizing enzyme, phenylethanolamine N-methyltransferase (PNMT). Embryonic MO was explanted from E13 or E18 embryos and maintained for up to 3 weeks. Successful culture of adrenergic neurons was possible only in explants taken from young embryos, since E18 explants failed to develop. In E13 explants, immunoreactivity to both PNMT and tyrosine hydroxylase, the rate limiting enzyme in catecholamine synthesis, was observed. PNMT catalytic activity which was barely detectable at the time of explanation increased markedly during the first week in vitro. To study the effects of glucocorticoids on PNMT development in central neurons, MO explants were grown in glucocorticoid deficient medium in which rat serum from adrenalectomized rats was substituted for human placental serum. Addition of natural glucocorticoids, cortisol or corticosterone, or the mineralcorticoid, deoxycorticosterone, during the third culture week had no effect on PNMT activity. Dexamethasone (DEX), a synthetic glucocorticoid, also had no effect on PNMT during the first or second weeks in culture. However, addition of DEX during the third culture week resulted in a doubling of PNMT activity. However, attempts to block the DEX effect during the third week or to block the increase in PNMT activity during the first week in control cultures with the glucocorticoid receptor antagonist, dexamethasone 21-mesylate, were unsuccessful. These results suggest that PNMT in central neurons does not require glucocorticoids for ontogeny during the embryonic period. This is in contrast to PNMT in adrenal medulla which requires glucocorticoids for normal development during both the embryonic and postnatal periods. More generally, these studies suggest that development of the same neurotransmitter phenotype in brain and periphery may be differentially regulated.