Anterograde and retrograde tracing with biotinylated dextran amine and Phaseolus vulgaris leukoagglutinin was used to assess projection patterns within the vibrissae representation of the rat's primary somatosensory cortex (S-I). Large and small injections of either tracer into the center of the vibrissae representation yielded dense anterograde and retrograde labelling throughout much of the tangential extent of the vibrissae representation within S-I. In all layers, the pattern and extent of retrograde and anterograde label was in rough congruence. The organization of this labelling varied across cortical layers. In layers II and III, labelled fibers extended away from injection sites in all directions and yielded a uniform pattern, which decreased in density with increasing distance from the tracer injection. There was a tendency for labelling to be more extensive along the representation of the row of vibrissae follicles that included the injection site than across rows. There was also a tendency for anterograde labelling to be more extensive in the direction of the representation of follicles more rostral on the face than that injected. In lamina IV, both labelled fibers and cells were restricted for the most part to the septa regions between the barrels. However, a small number of retrogradely labelled neurons were also located in the barrels (approximately one-ninth of the number found in the septa). The pattern observed in laminae II-III was repeated in layers V and VI. In these laminae, there was no evidence of a pattern of intracortical connections related to the vibrissae representation in overlying lamina IV.

New Zealand white rabbits were used to determine whether the changes in the Vth cranial nerve sensory root after compression were associated with the loss of a specific subclass of Vth cranial nerve ganglion cells, the disappearance of a distinct subset of primary afferent terminals in Vth cranial nerve nucleus caudalis, and/or injury to a specific axonal fiber type. There was no significant difference in the size of surviving ganglion cells after Vth cranial nerve compression, as measured 2 to 3 months after injury (P > 0.5, n = 4). Densitometric analysis of the nerves of rabbits that survived > 2 months after compression showed no significant difference in the immunoreactivity of substance P and calcitonin gene-reactive protein between compressed and control sides (P > 0.1, n = 4). Fink-Heimer staining of the Vth cranial nerve subnucleus caudalis revealed that transganglionic degeneration was most dense in the deeper layers, which are the sites of termination of large myelinated fibers. Ultrastructural evaluation of the type of myelinated axons injured by Vth cranial nerve compression in rabbits killed 7, 14, 37, and 270 days after injury was studied, and morphometric analysis was performed. The frequency distribution of axon diameters was significantly different for injured and control areas. The injured areas had higher ratios of small (< 3-microns diameter) to large-diameter axons compared to control distribution. These data indicate that balloon compression results in loss of fibers from the Vth cranial nerve sensory root and extensive transganglionic degeneration in the Vth cranial nerve brain stem complex. Cell size measurements and immunocytochemical data suggest that there is no specific loss of small ganglion cells or fine-caliber primary afferents. These experiments suggest that balloon compression relieves trigeminal pain by injuring the myelinated axons involved in the sensory trigger to the pain.

No abstract

Studies employing axoplasmic transport techniques have suggested that the central arbors of vibrissae-related primary afferents are rapidly and permanently lost from the trigeminal (V) brainstem complex after transection of the intraorbital nerve (ION). The present study reexamined this issue using immunocytochemistry for galanin (GAL) and anterograde labelling with Di-I to evaluate V brainstem organization in rats that sustained damage to the ION or individual vibrissae follicles in infancy or adulthood. After adult nerve damage, GAL-positive fibers are increased in layers I and II of V subnucleus caudalis (SpC). This was apparent by 3 days after the lesion. In rats that sustained nerve damage at birth (P0), GAL immunoreactivity (IR) appeared throughout the V brainstem complex and had a patchy distribution similar to that of vibrissae-related V primary afferents in normal rats. Increased GAL-IR in rostral portions of the V brainstem complex was observed in rats that sustained ION damage as late as P14. Additional experiments in which nerve damage was followed by destruction of the V ganglion demonstrated that this GAL-IR was contained in primary afferents. Damage to single vibrissa follicles or to a row of follicles produced a single patch or row of GAL-IR terminals in the somatotopically appropriate portion of the ipsilateral V brainstem complex. Di-I labelling in neonatally nerve-damaged rats demonstrated that primary afferent axons filled the central territory normally innervated by this nerve and that their terminal distribution was patchy. These results suggest that the V ganglion cells that survive neonatal axotomy may retain somatotopically organized projections to the V brainstem complex for at least a limited postnatal period.