Regulation of peptide neurotransmitter metabolism was examined in dissociated cell cultures of neonatal rat sympathetic and sensory ganglia. Previous studies have shown that pineal gland conditioned medium (PCM) influences substance P (SP) and somatostatin (SS) metabolism in sympathetic neurons in vitro. The present study examines mechanisms mediating these effects, and compares the actions of PCM on sympathetic and sensory neurons. PCM treatment increased SP levels in a dose-dependent manner without altering SS content of sympathetic neurons cultured in the presence of ganglion non-neuronal cells. Conversely, treatment of pure sympathetic neuron cultures resulted in a dose-dependent increase in SS, while SP was virtually undetectable at all doses. By contrast, dorsal root ganglion, trigeminal ganglion, and nondose ganglion sensory neurons contained SP both in the presence and absence of ganglion non-neuronal cells. Moreover, in each of these neuronal populations treatment with PCM increased SP levels both in the presence and in the absence of ganglion non-neuronal cells. These observations suggest that ganglion non-neuronal cells are necessary for sympathetic but not sensory neuron expression of SP. Moreover, PCM apparently stimulates SP in neurons which already contain the peptide, but the factor cannot foster de novo expression of the phenotype. PCM also influenced other transmitter traits in sympathetic neurons, suggesting linkage between mechanisms regulating peptides and other transmitters. In cultures containing both sympathetic neurons and non-neuronal cells, PCM treatment increased cholineacetyltransferase (CHAC) activity as well as SP, and decreased tyrosine hydroxylase (TOH) activity. By contrast, PCM treatment of pure sympathetic neuron cultures led to parallel increases in SS and TOH activity with negligible levels of SP and CHAC. These observations suggest that in sympathetic neurons, SS may be linked with noradrenergic expression, while SP is associated with cholinergic development, although more data are required to confirm this relationship. Moreover, there may be a reciprocal relationship between SP and SS expression by sympathetic neurons analogous to previous observations regarding cholinergic-noradrenergic expression (P. H. Patterson and L. L. Y. Chun, Proc. Natl. Acad. Sci. USA 71, 3607-3610, 1974; Dev. Biol. 56, 263-280, 1977). Consequently, neurotransmitter phenotypic expression is a complex process in which the environment regulates a balance among multiple transmitters.