Quantitative study of the tectally projecting retinal ganglion cells in the adult frog. II. Cell survival and functional recovery after optic nerve transection.

It is known from previous work that ganglion cells disappear from the retina in significant numbers during optic nerve regeneration in the adult frog. In the present study, the population size of surviving ganglion cells that have returned axon terminals to the correct tectal loci was estimated by counts of retrogradely labeled cells in retina-flat-mounts after tectal injections of HRP. Bilaterally symmetric injections were delivered to allow comparison of the normal and affected retinas. The frogs studied had regenerated the left optic nerve and had visually guided behaviors initiated by the recovered eye (see below). The proportion of tectally projecting ganglion cells in the normal retinas and in retinas of normal frogs studied in parallel ranged from 83-86% (Singman and Scalia: J. Comp. Neurol. 302:792-809, 1991). In the affected retinas, the subpopulation of tectally projecting cells was reduced by 40-90% after regeneration, and the relative size of this subpopulation ranged from 67-86%. The optic tectum was injected unilaterally in one specimen, on the side ipsilateral to the regenerated (left) optic nerve. The HRP-labeled ganglion cells in the ipsilateral (left) retina accounted for only 0.8% of the surviving ganglion cells in this animal, whereas it was previously found that the ipsilateral tectally projecting ganglion cells normally amount to 0.9-2.3% (Singman and Scalia, op. cit.) In frogs recovering from transection of the left optic nerve, the frequency, latency, and accuracy of the prey-acquisition responses initiated by the recovering eye were compared with those initiated by the normal eye. Mealworms or lure dummies were used to stimulate prey acquisition. In one experiment, the stimuli were presented unilaterally in the monocular fields of frogs permitted to use both eyes. Prior to the fourteenth postoperative week, the affected eye initiated responses of abnormally long latency and low frequency. In contrast, responses initiated by the affected eye after 14 weeks appeared to be normal in all respects. In another experiment, the normal eye was sutured shut in some frogs recovering for at least 24 weeks and then the affected eye was retested. The affected eye was capable of consistently initiating brisk and accurate prey acquisition. In a final experiment, two stimuli were presented simultaneously in bilaterally symmetric regions of the monocular fields of frogs surviving at least 42 weeks. These fully recovered frogs showed no preference for using either the normal or the recovered eye. Despite severe loss of tectally projecting ganglion cells during optic nerve regeneration, frogs are capable of apparently normal visual responses in prey acquisition tests.