Quantitative ultrastructure of synapses on functionally identified primary afferent neurons in the cat trigeminal mesencephalic nucleus

Abstract.

Though a number of studies have reported the presence of synapses on neurons in the trigeminal mesencephalic nucleus (Vmes), there have been no quantitative studies of either the density of innervation, or the ultrastructure, of the synapses on single, physiologically identified neurons in this nucleus. In this study we recorded from single neurons in the Vmes, identified them as being either muscle spindle afferents (MS) or periodontal ligament mechanoreceptor afferents (PL), and then labeled the neurons by intra-axonal injection of horseradish peroxidase (HRP). The material was first processed to reveal the HRP activity, following which ultrathin sections through the labeled somata were cut and examined under the electron microscope. Complete serial reconstructions were made through the soma of one MS neuron and one PL neuron, and the contacts on the neurons reconstructed. Boutons were found on the soma, spines, appendages and the axon hillock and the initial segment of the axon. The numbers of boutons terminating on the two neurons were 198 (PL) and 424 (MS), giving a packing density of 4.4 and 10.7 boutons respectively (i.e., number of boutons/100 µm² of the postsynaptic membrane). Boutons could be separated into two types on the basis of their vesicles: those containing clear, round vesicles (i.e., S-type) and those containing a mixture of round, oval and flattened vesicles (P-type). Ninety-five (PL neuron) and 99% (MS neuron) of terminals on the two neurons were P-type. All the S-type boutons and 80% of the P-type boutons formed asymmetric synaptic contacts while 10% of the P-type boutons made symmetric contacts. Quantitative measurements of the P-type boutons on the labeled neurons, in which the data of MS and PL neurons were pooled, revealed that bouton volume was highly correlated with bouton surface area, active zone number, total active zone area, vesicle number, and mitochondrial volume. However, comparing the quantitative measurements of the P-type boutons with those of previously reported vibrissa afferent terminals and their associated axon terminals revealed that all the parameters were smaller for the P-type boutons (on Vmes neurons) than those of the vibrissa afferent terminals but similar to those of axon terminals presynaptic to the vibrissa afferents. Taken together, our results emphasize the wide scope for synaptic interactions in the Vmes and suggest that it may be more fruitful to view the Vmes as an integrating center.