ELECTRON-MICROSCOPIC EVIDENCE FOR COEXISTENCE OF LEUCINE(5)-ENKEPHALIN AND GAMMA-AMINOBUTYRIC-ACID IN A SUBPOPULATION OF AXON TERMINALS IN THE RAT LOCUS-COERULEUS REGION

We recently described ultrastructural evidence for morphologically het erogeneous axon terminals containing the endogenous opioid peptide, me thionine(5)-enkephalin (ENK), that formed synapses with neurons contai ning the catecholamine synthesizing enzyme, tyrosine hydroxylase, in t he locus coeruleus (LC) of the rat brain. The morphological characteri stics of these terminals suggested that ENK may be co-localized with e ither an excitatory or inhibitory amino acid. To further test this hyp othesis, we combined immunogold-silver localization of gamma-aminobuty ric acid (GABA) and immunoperoxidase labeling for ENK in single sectio ns through the LC, in the present study, to determine whether ENK and GABA were contained within single axon terminals. Light microscopic an alysis of ENK and GABA immunoreactivities in the LC indicated that bot h transmitters were enriched in the dorsal pens. Although electron mic roscopy revealed that ENK and GABA were located primarily in axon term inals, some dendrites also contained immunolabeling for GABA. The dens e core vesicles were consistently the most immunoreactive in ENK conta ining axon terminals and were identified toward the periphery of the a xon terminal distal to the synaptic specialization. Axon terminals con taining either ENK or GABA immunoreactivities contained pleomorphic ve sicles as well as large dense core vesicles, varied in size and formed heterogeneous types of synaptic specializations (i.e. asymmetric vs. symmetric). Approximately 38% (n = 76) of the axon terminals containin g ENK immunoreactivity (n = 200) also contained GABA. Some axon termin als containing peroxidase labeling for ENK (22%; n = 44) converged on common targets with GABA-labeled axon terminals. Finally, a few ENK-la beled axon terminals (14%; n = 28) formed asymmetric (excitatory-type) synapses with dendrites containing gold-silver labeling for GABA. The results, therefore, indicate that the opioid peptide, ENK, and the in hibitory amino acid, GABA, may influence LC neurons by concerted actio ns via (1) release from a common axon terminal, and (2) via separate s ets of afferents converging on similar portions of the plasmalemma of target neurons. Furthermore, these studies also suggest a cellular sub strate for opioid inhibition of LC neurons via activation (i.e. asymme tric synapses) of inhibitory GABAergic neurons. Future studies are req uired to determine whether the receptive sites for ENK and GABA are lo cated at similar sites on the plasma membranes of LC neurons pre- or p ostsynaptically and whether there is differential release of either tr ansmitter from single terminals in the LC.