Neuropeptide genes are expressed in specific subsets of large polyploid neurons in Aplysia californica. We have defined the transcription initiation sites of three of these neuropeptide genes (the R14, Lll, and ELH genes) and determined the nucleotide sequence of the promoter regions. The genes contain the usual eucaryotic promoter signals as well as other structures of potential regulatory impoortance, including inverted and direct repeats. The Lll and ELH genes, which are otherwise unrelated, have homology in the promoter regions, while the R14 promoter was distinct. When cloned plasmids were microinjected into Aplysia neurons in organ culture, transitions between supercoiled, relaxed circular, and linear DNAs occurred along with ligation into high-molecular-weight species. About 20% of the microinjected neurons expressed the genes. The promoter region of the R14 gene functioned in expression of the microinjected DNA in all cells studied. When both additional 5' and 3' sequences were included, the gene was specifically expressed only in R14, suggesting that the specificity of expression is generated by a multicomponent repression system. Finally, the R14 peptide could be expressed in Lll, demonstrating that it is possible to alter the transmitter phenotype of these neurons by introduction of cloned genes.Control of gene expression in the central nervous system (CNS) may require novel and complex regulatory networks to generate the vast and intricate communication network between cells. The specificity of this network is determined by the expression of specific sets of genes, including those which determine neurotransmitter and receptor phenotype. In addition, the CNS can be considered to develop continuously throughout the life of an animal as experience shapes the fine structure of this communication network. The complexity, which affords the nervous system with remarkable properties, makes cellular and molecular studies, including those of gene expression, a difficult task. Neurons are generally postmitotic, and the vast number of unique cells makes the study of identified neurons with known functions very difficult in vertebrates.Many invertebrate species, including mollusks, have provided partial solutions to these problems due to the numerical simplicity of their CNSs and the large size of the component neurons.