Ultrastructural distribution of 125I-toxin F binding sites on chick ciliary neurons: synaptic localization of a toxin that blocks ganglionic nicotinic receptors.

Although alpha-bungarotoxin (BGT), a common probe for nicotinic ACh receptors from vertebrate skeletal muscle, binds tightly to many autonomic ganglia, it fails to block nicotinic transmission in most of these ganglia. Recently, we have isolated a second toxin, toxin F, that blocks transmission in several autonomic ganglia, including the chick ciliary ganglion. 125I-Toxin F binds to 2 sites in the ciliary ganglion: one site that is also recognized by BGT and one site that is not. Since the presence of BGT fails to prevent the blocking effect of toxin F, the toxin F site not recognized by BGT most likely represents the neuronal nicotinic receptor. Accordingly, we have localized the binding of 125I-toxin F to both sites, using electron-microscopic autoradiography. After a 45 min incubation, 125I-toxin F binding sensitive to BGT was primarily localized extrasynaptically on the neuronal plasma membranes; however, by 4 hr, much of this site had been internalized. In contrast, the 125I-toxin F binding site not recognized by BGT was highly concentrated near synaptic membranes at both times. Pretreatment of ganglia with the classic nicotinic antagonists dihydro-beta-erythroidine (DHBE) and d-tubocurarine (DTC) significantly reduced 125I-toxin F binding to the toxin F-specific site. The average density of these sites on synaptic membranes was approximately 600 sites/micron 2. We conclude that toxin F binds to 2 pharmacologically distinct sites in the ciliary ganglion and that these sites are distributed differently over the plasma membrane of ciliary neurons. On the basis of the density of the toxin F-specific binding sites at synaptic membranes, we infer that the density of synaptic nicotinic receptors on these neurons is at least 20-fold lower than the density of nicotinic receptors at the vertebrate neuromuscular junction, as determined by BGT binding. These findings are consistent with those of electrophysiological studies, which also suggest low nicotinic receptor densities on ganglionic neurons.