Quantitative electron microscopic evidence for reinnervation in the adult rat interpeduncular nucleus after lesions of the fasciculus retroflexus

Abstract

The method of electron dense degeneration has been used to make a quantitative study of the projection from the habenula through the fasciculus retroflexus (FR) to the interpeduncular nucleus (IPN) in the rat. The IPN is a midline structure onto which the right and left fasciculi converge. In the rostral part of the IPN the fascicular axons from each side form synapses throughout the mediolateral extent of the ventral two‐thirds of the nucleus. In the caudal part of the IPN the fascicular axons from each side terminate to an equal extent in two discrete, parasagittal zones, situated one on each side in the mid‐mediolateral extent of the IPN. These zones contain clusters of neurons located along the course of a characteristic row of arterioles and venules penetrating the IPN from its ventral surface. In both rostral and caudal parts of the IPN the fascicular axons form single synaptic contacts with the dendrites of the interpeduncular neurons, but caudally, in the two parasagittal zones they also form crest synapses. Crest synapses are only found in this part of the IPN. In crest synapses two presynaptic terminals form markedly asymmetrical contacts with the parallel opposing sides of an attenuated dendritic appendage (the crest). After unilateral fascicular lesions only one member of a pair of axon terminals contacting a crest degenerates. After bilateral fascicular lesions, however, there are many instances in which both members of a crest pair degenerate. This indicates that the axon terminals from the right and left fasciculi are segregated at the level of the crests, in such a way that one terminal comes from the right fasciculus and one from the left. At longer survivals after unilateral or bilateral fascicular lesions the degeneration is completely removed, but crest synapses are still present, indicating that the presence of fascicular axons is not necessary for the maintenance of crests in the IPN. To investigate the effects of chronic deafferentation, the left fasciculus was destroyed and, after a survival of at least six weeks (sufficient for all degeneration to be removed) the right fasciculus was destroyed one day before killing. Under these conditions there are many crests in which both axon terminals show degeneration. The proportion of such doubly degenerating crest synapses is similar to that found after acute (1 day) bilateral lesions, indicating that axons from the right fasciculus have reinnervated sites formerly occupied by the left fasciculus. We conclude that during normal development there is some constraint which prevents both sides of a crest being innervated by axons from the fasciculus of the same side of the brain, but that this constraint is not effective after unilateral fascicular lesions in the adult.