We used isolated IgG antibodies selective for P2X3 receptors to study the ultrastructural distribution of these receptors in rat sensory neurons. In trigeminal ganglia, P2X3 receptor immunoreactivity occurred in small and large nerve cell bodies and their processes. Endoplasmic reticulum and Golgi apparatus were heavily stained; cytoplasmic matrix was faintly to moderately stained. In synaptic glomeruli in lamina II of cervical dorsal horn, P2X3 receptor-immunoreactive core terminals were postsynaptic to unlabelled vesicle-containing dendrites and axons. In the nucleus of the solitary tract, receptor-positive boutons synapsed on dendrites and cell bodies and had complex synaptic relationships with other axon terminals and vesiculated dendrites. These observations identify sites from which ATP could be released to influence sensory signalling within the central nervous system.

To develop a method for quantitative electron microscopic immunocytochemistry on neural tissue of CNS, we tested the extent to which ethanol treatment would improve the penetraaon of immunoreagents through vibratome sections fmed in high concentrations of glutaraldehyde without compromising ultrastructure. " v e r s e or sagittal vibratome sections (60-80 pm) of spinal cord perfused with 1% formaldehyde plus 1% or 2.5% glutaraldehyde wen washed in 50°/o ethanol for 0-70 min and stained to re-veal immunoreactivity for neuropeptide Y (NPY). Semi-thin (1 pm) or ultra-thin sections were used to assess the depth to which NPY nerve fibers in the dosal horn were stained. Without ethanol washing, immunoreactive nerve fibers were visualized only in the surface 5-10 pm of transverse or sagittal vibratome sections.In tr;uls7mse vibratome Sections, NPY nerve fibers, which ran perpendicular to the cut surfaces of the sections, were en&ely stained aftet a 30-& wash in 50% ethanol. The numbers of NPY-immunoreactive varicosities and synapses were comparable at the surfaces and in the centers of the vibratome sections. In sagittal sections, where NPY nerve fibers

Noradrenergic axons in the enteric plexuses of the guinea-pig ileum have been identified at the ultrastructural level using three techniques: the chromaffin reaction, localization of dopamine-fl-hydroxylase (DBH) with horseradish peroxidase-conjugated antibody, and in vivo and in vitro loading with 5-hydroxydopamine (5-OHDA).In the myenteric (Auerbach's) plexus from normal ileum all of these methods produced electron-dense deposits in a distinctive population of axonal varicosities that contained many flattened vesicles (usually more than 30% of the total number of vesicles), as well as oval or irregularly shaped vesicles. When noradrenergic axons to the small intestine had degenerated after surgical denervation, no profiles containing vesicles with electron-dense deposits were observed with the chromaffin reaction, DBH localization or loading with 5-OHDA. Pretreatment with 6-hydroxydopamine (6-OHDA) substantially reduced the number of noradrenergic axons identified by these three techniques. Axons with many flattened vesicles of similar dimensions but without dense cores were found in myenteric plexus from conventionally fixed intestine. These axons had the same distribution within the ganglia as cytochemically labelled noradrenergic terminals and disappeared after extrinsic denervation.In the normal submucous (Meissner's) plexus, both 5-OHDA loading and the chromaffin reaction produced electron-dense granules in small and large ve~icles in some axon terminals. In ganglia labelled by these techniques, reactive terminals contained many small round vesicles and few flattened and large round vesicles as did a population of nonreactive terminals. In axon terminals of submucous plexus labelled with anti-DBH, flattened vesicles were found to be more numerous than with the other treatments. As in the myenteric plexus, all reactive axons disappeared from the submucous plexus after extrinsic denervation. In conventionally processed submucous ganglia, noradrenergic axon profiles could not be distinguished from some non-noradrenergic profiles on the basis of types and proportions of vesicles.In the myenteric plexus noradrenergic axon terminals were seen most often near the edges of ganglia. Noradrenergic varicosities also occurred near nerve cell bodies but were rarely found in internodal strands. In the submucous plexus noradrenergic terminals appeared to be 0300-4864/81/020331-22504.20/0 9 1981 Chapman and Hall Ltd. 332LLEWELLYN-SMITH, WILSON, FURNESS, COSTA and RUSH randomly distributed throughout submucous ganglia. No axosomatic synapses formed by noradrenergic axons were found in either plexus, but synapses on nerve processes were occasionally encountered in submucous ganglia.

Electron microscopic immunocytochemistry has been used to study the distribution and synaptic relationships of substance P (SP)- immunoreactive nerve fibers in ultrathin sections from whole-mount preparations of myenteric ganglia from guinea pig small intestine. At the light microscopic level, myenteric ganglia stained for ultrastructural study contained dense arrays of SP-immunoreactive nerve fibers around and between the neuronal and glial cell bodies. At the electron microscopic level, SP-containing nerve fiber profiles occurred throughout the neuropil of the myenteric ganglia. Both vesicle- containing and nonvesiculated nerve fiber profiles were immunoreactive. The positive vesiculated profiles contained variable proportions of small clear and large granular vesicles. Two-thirds of the vesicle- containing nerve fiber profiles in myenteric ganglia were immunoreactive for SP. Many vesiculated SP-immunoreactive nerve fiber profiles were directly apposed to each myenteric neuron. However, only about 0.6% of these vesiculated profiles formed synapses that showed the ultrastructural specializations of vesicle clustering presynaptically and fuzzy electron-dense material on the postsynaptic membrane. On the other hand, morphologically identifiable synapses with SP-immunoreactivity comprised about half the total number of synapses in the ganglia. SP-immunoreactive synapses were observed on nonvesiculated nerve processes in the periphery of the ganglia and on nerve cell bodies. Most of the axosomatic SP synapses occurred on small neurons that lay either on the surfaces of the ganglia or near the fibers of the internodal strands where they traveled through the ganglia. Both SP-positive and SP-negative nerve cell bodies received SP synapses.(ABSTRACT TRUNCATED AT 250 WORDS)

No abstract