Recovery from conduction failure in optic axons spared by lesions in the rat.

A chronic preparation has been developed to determine the physiological sequelae of a lesion of optic axons, and to correlate these with the optic innervation of the superior colliculus (SC). During deep pentobarbital anesthesia: (1) Vigorous small-field (2-5 degrees) multi-unit responses were recorded from microelectrodes in the lower stratum griseum superficiale (SGS) and upper stratum opticum (SO), evoked by the motion of a 40 degrees black edge. (2) The maps of the retinal projection, obtained without immobilization of the eye, were regular in about 1/3 of cases and only slightly irregular in most of the remainder. (3) There was a rhythmic (10-20/s), retina-dependent firing of spike bursts in visually responsive SC. At a lighter level of anesthesia this bursting reduced or disappeared, and coincidentally there was a rapid enlargement of the multi-unit receptive fields (MU-RFs). It is suggested that GABA-based inhibitory mechanisms were involved in producing both the rhythmic bursting and the smaller-field MU-RFs. At either anesthetic level the movement of a 5 degrees disc produced 20-65 degrees MU-RFs. Vertical lesions which severed part of the visual pathway reduced and, if large enough, abolished responses even to a strong testing stimulus that normally produced a response from an area of visual field larger than the range of eye movement. The region of SC affected was always medial or lateral, and sites within the still-responsive area had normally small MU-RFs. Repeated mappings over the following hours to days revealed a variable recovery of response; this recovery always occurred within the silenced zone of SC nearest to the still-responsive area. Ultimately two discrete zones, responsive and silent, persisted. Mapping the SC, at 500 microns intervals, involved about 25 microelectrode penetrations. The number of sites silenced immediately postlesion was 1-24. Response returned at 1-15 of these, corresponding to 4-60% of the entire SC. Most of the recoveries occurred by 2 h postlesion, but occasionally an additional 1-3 sites recovered during the following 2-8 days. Recovered MU-RFs were small, and appropriately located within the map. Intraocular horseradish peroxidase (HRP) accumulated in severed optic axons but was transported with no obvious hindrance in spared axons alongside the lesion. Within the SC a sparse optic innervation extended beyond the responsive area, into about 0.5 mm of the region which remained silent after the recoveries were completed.(ABSTRACT TRUNCATED AT 400 WORDS)