Plasticity of retinofugal projections after partial lesions of the retina in newborn syrian hamsters

Abstract

AbstractIn Syrian hamsters, retinofugal terminal projections are just beginning to form on the day of birth; on this day, unilateral partial lesions of the retina were produced alone or in combination with one, two, or all of the following procedures: enucleation of the opposite eye, bilateral ablation of the posterior neocortex, transection of the contralateral brachium of the superior colliculus (bsc). When the animals were young adults, the projections of one or both eyes were determined with experimental neuroanatomical methods.Following retinal lesions alone, the zones in the superior colliculus (SC), dorsal lateral geniculate nucleus (LGd), and ventral lateral geniculate nucleus (LGv) corresponding to the reconstructed area of ganglion cell destruction in the opposite eye may be totally, partially, or not at all filled in by the axons of surviving ganglion cells in the contralateral retina.Independent of the size of the retinal lesions, the surviving retinal areas showed a variable degree of ganglion cell loss. In the SC the degree of filling‐in of the zones corresponding to the retinal lesion is correlated with the number of surviving ganglion cells in the areas not directly damaged by the early eye surgery. The same appears to be true for the LGd and LGv, but for these structures our quantitative assessments are less accurate. This finding implies that the total number of retinofugal axons reaching these structures may be the critical parameter determining the filling‐in of the areas corresponding to the early retinal lesion. Such a result may be explained if interactions among optic axons influence their distribution within those structures in which they terminate, e.g., growing axon tips may “repel” each other by a kind of contact inhibition of growth. Species differences in the consequences of early partial retinal ablation may be due to variations in the effects of surgery on the number of ganglion cells surviving in the remainder of the eye, and not to differences in the mechanisms governing the formation of retinotopic connections. However, the possibility that our, and other, results are due to varying degrees of regulation in the retina cannot be ruled out by the present findings.In animals subjected to neonatal retinal lesions, one can trace an anomalous projection from the intact eye to the ipsilateral stratum griseum superficiale (SGS). However, the filling of the zone corresponding to the retinal lesion is subject to constraints in addition to those which may be imposed by anomalous projections to that region from the intact eye, or from the visual cortex; in animals subjected at birth to the destruction of one or both of those afferent systems, the variability in the filling of the zones corresponding to the retinal lesion did not appear to be altered. The plasticity of the retinotectal projection is also unaffected when the SC is partially denervated by transection of the bsc at the time of the retinal lesion.In normal hamsters, the full thickness of the SGS is filled by the terminal arbors of axons originating from ganglion cells in the contralateral eye. In hamsters subjected to neonatal retinal lesions, there are regions in which the contralateral projections of the operated eye are absent or anomalously sparse in the deep half of the SGS. This in interpreted as a tendency for crossed retinofugal axons to preferentially arborize close to the surface of the SC. Similar phenomena are observed in the LGd and one of the laminae of the LGv. On the basis of this and the proposed interactions among retinofugal fibers, specific predictions can be made regarding the morphology of retinal ganglion cell axonal arbors in our experimental animals.