Visual information is carried from eye to brain by retinal ganglion cells, whose fibers form the optic nerve. Different types of ganglion cells receive messages about the visual world via the various types of amacrine cells and bipolar cells that make functional contact with them. They convey these messages to a number of different sites in the brain, where they subserve a variety of distinct visual functions. Although some ganglion cell types have been extensively studied, most have not. Furthermore, even for those that are best studied, we know almost nothing of the different types of amacrine and bipolar cells that contact them. The long-term objectives of these studies are to elucidate the properties of each of these types of ganglion cell. Such an understanding is fundamental to an appreciation of clinical correlates of malfunction. For example, although we know that optic nerve compression has a differential effect on optic nerve fibers of different diameters, we can say little at present about the sequence of functional deficits that could be expected to result from progressive compression of the nerve. Four specific projects are proposed in this application. The first is to characterize the different ganglion-cell types using morphologic criteria combined with knowledge of the central projection of the cell. The study is based upon an in vitro preparation, developed in this laboratory, in which ganglion cells labelled by a central injection of a non-toxic fluorescent dye, are injected, under direct microscopic visualization, with a substance (horseradish peroxidase) that allows their detailed morphology to be revealed. Large samples of well-labelled cells with known central destinations can be collected in this manner, and used to clarify both the complement of ganglion cell types found in the retina, and the central destinations of these types. The second project involves determining the functional properties of each of these morphologically characterized types. It proves possible to identify fluorescence labelled ganglion cells in the living eye, using a new technique. In regions of relatively low ganglion-cell density, these cells can be recorded from intra-ocularly, using an established technique. Once the functional properties of a number of labelled cells have been determined, and the retinal positions of these cells marked, they can then be intracellularly injected with horseradish peroxidase, using the methods of the first project. The third project deals with inputs to the ganglion cells, and attempts to determine which types of amacrine and bipolar cells contact a given type of ganglion cell. The method involves a double-label in vitro experiment in which an amacrine or bipolar cells contact a given type of ganglion cell. The method involves a double-label in vitro experiment in which an amacrine or bipolar cell is injected together with an overlying ganglion cell. When sites of contiguity are determined between the processes of the two cells, these sites will be cut from the tissue, sectioned, and viewed with the electron microscope to determine if these sites of apparent contact are synaptic. The fourth project involves the fine structure and possible functional contacts of what we term 'axonal twigs'. These are processes up to 10 (mu)m in length found on the axons of most primate ganglion cells, as well as some cat ganglion cells. They are found within a few hundred micrometers of the soma, in number up to about 20, and are directed laterally or sclerally. We propose to investigate the fine structure these processes using the methods described for the third project.
