Aromatic L-amino acid decarboxylase (AADC, EC 4.1.1.28) catalyzes the decarboxylation of L-dopa to dopamine in catecholamine cells and 5-hydroxytryptophan to serotonin in serotonin-producing neurons. This enzyme is also expressed in relatively large quantities in nonneuronal tissues such as liver and kidney, where its function is unknown. Neuronal and nonneuronal tissues express AADC mRNAs with distinct 5' untranslated regions. To understand how this is accomplished at the genomic level, we have isolated rat genomic DNA encoding AADC. The organization of the AADC gene suggests that there are two separate promoters specific for the transcription of neuronal and nonneuronal forms of the AADC message. A small exon containing 68 bases of the neuronalspecific 5' end is located -9.5 kilobases upstream of the translation start site, which is contained in the third exon. Approximately 7 kilobases upstream from the neuron-specific promoter is another small exon containing 71 bases ofthe 5' end of the nonneuronal AADC message. These data suggest that transcription initiating at distinct promoters, followed by alternative splicing, is responsible for the expression of the neuronal and nonneuronal forms of the AADC message.

The rat aromatic l‐amino acid decarboxylase (AADC) gene contains alternative promoters directing expression of neuronal and nonneuronal mRNAs that differ only in their 5′ untranslated regions (UTRs). We have analyzed the expression of the neuronal promoter of the AADC gene in cells synthesizing catecholamines and serotonin, as well as in non‐AADC‐expressing cells. We demonstrate the use of the neuronal‐specific UTR in individual dopamine‐, norepinephrine‐, and serotonin‐containing neurons. Transfection analyses show that the rat AADC neuronal promoter, containing 2,400 bp upstream of the transcription start site and including the 68‐bp untranslated exon 2, can activate transcription from a reporter gene in both catecholaminergic and serotonergic cell lines. These analyses identified several positive and negative cis‐active elements within this region. Unexpectedly, we observed that this promoter, when removed from its native context within the AADC gene, can also direct expression of a reporter gene in cells that do not normally express AADC mRNA. These results suggest that tissue‐specific expression of the neuronal promoter may not be controlled by cis‐active elements within the first 2,400 bp of the promoter. Additional information may be required to restrict neuronal promoter expression to appropriate cell types. This regulatory information could reside elsewhere within the promoter, within introns, or may be provided by interactions between the two AADC promoters.