Search

Orthograde and retrograde axoplasmic transport have been studied in the optic nerve heads of 37 Macaca fascicularis eyes with normal or elevated intraocular pressure (IOP) produced by cannulation of the anterior chamber. Orthograde transport was labeled by 3H-amino acids injected intravitreally and incorporated into retinal ganglion cell proteins. Retrograde transport was studied in the same eyes by injecting horseradish peroxidase (HRP) into one or both optic tracts and dorsal lateral geniculate nuclei (dLGN). Both tracers accumulated in the lamina scleralis (LS) of eyes maintained at pressures of 25 to 150 mm. Hg for 12 to 28 hours (pressure in normal controls = 10 to 14 mm. Hg) but the HRP technique was markedly more sensitive. The degree of retrograde transport obstruction in the LS appeared to be directly proportional to both the height and the duration of elevated IOP. In one experiment, the blockades of orthograde and retrograde transport induced at 50 mm. Hg were demonstrated to be reversible. Serial reconstructions of radioautographs and peroxidase-reacted sections of the optic nerve heads demonstrated that the orthograde and retrograde transport obstructions were coincidental anatomically by light microscopy in the LS and occurred most prominently in the temporal quadrants of the nerve head. These transport obstructions occurred at moderate elevations of IOP (25 TO 50 mm. Hg) despite (1) elevated arterial PO2 levels during inhalation of 100 percent oxygen and (2) intact nerve head capillary circulation, as demonstrated by perfusion with nucleated avian erythrocytes.

In the primate, ganglion cells of the temporal retina project ipsilaterally and those of the nasal retina, contralaterally into the optic tract. The vertical meridian passing through the fovea defines the border between these two populations of ganglion cells and has been demonstrated in four Macaque monkeys after unilateral injection of horseradish peroxidase into the dorsal lateral geniculate nucleus and examination of the pattern of retrograde labeling of those ganglion cells projecting to the injected side. A median 1 degree vertical strip in which ipsi- and contralaterally projecting ganglion cells intermingle was found, confirming the report by Stone et al. ('73). In addition, occasional extrafoveal labeled ganglion cells were found as far as 2 degrees from the vertical midline in the otherwise unlabeled hemiretinae. These ganglion cells were not numerous and had somata of all sizes, suggesting that they do not constitute a separate class of ganglion cells as found in the temporal retina of the cat. In contrast to the description by Stone et al. ('73), the strip of vertical overlap did not show a constant width through the fovea, since mixing of labeled and unlabeled ganglion cells was found in a band approximately 1/2 degree wide along both the nasal and temporal rims of the foveal pit which is 500 mum (2 degrees) in diameter. Beyond these 1/2 degree arcs, the appropriate hemiretina was either completely unlabeled, or contained virtually every ganglion cell labeled on the side projecting to the injected dorsal lateral geniculate nucleus. The scattered labeled ganglion cells rimming an otherwise unlabeled hemifovea represent a possible anatomical basis for the phenomenon of "macular" or "foveal sparing" in which unilateral damage to the occipital cortex produces homonymous hemianopsia with sparing of a small island of centralmost vision extending about 1 degree from the foveal center. From this study, it is not possible to define the receptive fields or specific photoreceptor connections of the ganglion cells labeled with horseradish peroxidase, so that at the present time quantitative correlations cannot be made between the numbers of ganglion cells remaining on the affected side of the fovea and the extent of preservation of visual function in the spared zone. The presence of labeled ganglion cells rimming the fovea in its entirety is compatible with the sequence of foveal development in late prenatal life. After lateral displacement both nasally and temporally of ganglion cells which initially lay in the median vertical overlap strip of 1 degree, in the adult retina a strip approximately 1/2 degree wide around the perimeter of the foveola should contain a mixture of ipsi- and contralaterally projecting ganglion cells. The total population of ganglion cells beyond this 1/2 degree band should be completely ipsi- or contralateral in their projection patterns, as is observed...