Ultrastructural correlates of transmitter release in presynaptic areas of lamprey reticulospinal axons

Abstract

The ultrastructure of presynaptic areas of lamprey reticulospinal axons was studied before, during, and after periods of elevated transmitter release produced either by repetitive action potential activity or depolarization by elevated extracellular potassium. Controls for possible effects of these procedures per se were done by replacing extracellular Ca with Mg to block transmitter release. In some experiments the time course of ultrastructural changes during K depolarization and subsequent recovery were studied by fixing tissue samples at various times. Transmitter release produced by action potential activity (20/sec for 15 min) in the presence of extracellular Ca significantly and reversibly decreased the number of synaptic vesicles, the area occupied by the vesicles, and the density of synaptic vesicles. An unexpected finding was a reversible decrease in the length of the differentiated membrane during periods of increased transmitter release. Transmitter release significantly and reversibly increased the number of coated vesicles, expanded the presynaptic membrane, and increased the number of pleomorphic vesicles. K depolarization (50 mM K for 15 min) produced identical, reversible effects, except that the expansion of the presynaptic membrane, although significant, was relatively small and there was no change in the number of pleomorphic vesicles. Raising the temperature of the saline from 2°C (K depolarization experiments) or 7° C (action potential experiments) to 20° C did not change the results qualitatively but did produce somewhat larger effects during stimulation and appeared to increase the speed of recovery. Action potential activity or K depolarization in control experiments with the Ca in the saline replaced by Mg had little or no effect on synaptic ultrastructure. Synaptic vesicles in lamprey reticulospinal axons never contacted the axonal membrane anywhere other than at the differentiated membrane. During periods of elevated transmitter release, although the absolute number of vesicles in contact with the differentiated membrane decreased, the percentage of total vesicles in contact with the differentiated membrane increased dramatically. This suggests that (1) the differentiated membrane is the site of vesicle release and (2) there is an active process of vesicle movement to this membrane. In the course of this work it was observed that presynaptic areas closer than approximately 2 mm to the site of axonal transection, regardless of the composition of the saline or the experimental conditions, showed ultrastructural changes typical of increased transmitter release. The changes were profound after only several minutes following transection of the axon and were progressively greater for synaptic areas nearer the cut end. Intracellular recording from transected axons confirmed that their cut ends do not seal for at least 24 hr after transection, and, therefore, nearby synapses are depolarized and exposed to the extracellular fluid. These results from vertebrate central nervous system synapses suggest that during transmitter release synaptic vesicles are drawn to and fuse with the presynaptic membrane, causing it to expand, and are 'recycled' in a process requiring minutes, possibly involving coated vesicles and pleomorphic vesicle formation. The differentiated presynaptic membrane (active zone) appears to be the actual site of vesicle fusion, and its overall length is reversibly decreased by release, suggesting a transient depletion of some specialized membrane component.