Immunolocalization of cellular retinoic acid binding protein to müller cells and/or a subpopulation of GABA‐positive amacrine cells in retinas of different species

Publisher Website
Google Scholar
Abstract
Retinoic acid (RA) and its specific binding protein, cellular RA binding protein (CRABP), are found in relative abundance in bovine and rat retinas. Since RA does not participate in the visual cycle, the presence of RA and its binding protein in retina suggests that they may be involved in other aspects of retinoid action. As an initial step in identifying the role of RA and its binding protein in retina, monoclonal antibodies were prepared against CRABP purified from bovine retina and used to localize this antigen by immunocytochemistry in retinas of different species. Human and monkey retinas showed specific cytoplasmic labeling of Müller cells. Cat, bovine, rabbit, rat, turtle, and chick retinas showed specific cytoplasmic labeling of some somata in the inner nuclear and ganglion cell layers and characteristic strata in the inner plexiform layer. Cat and bovine retinas also showed cytoplasmic labeling of Müller cells. Immunoreactivity in these species was absent with nonimmune serum or abolished when the antibodies were preabsorbed with purified antigen. Chameleon, goldfish, and frog retinas were nonreactive. We used double‐labeling immunofluorescence experiments to determine if the CRABP‐positive cells were also positive for known neurotransmitters or associated enzymes. CRABP‐positive amacrine cells of cat, cow, rabbit, rat, and chick represented a subset of the more numerous γ‐aminobutyric acid (GABA)‐positive amacrine cells. However, turtle CRABP‐positive amacrine cells were negative for GABA despite the fact that turtle retina contains many GABA positive cells. CRABP‐positive amacrine cells in rat retinas were not immunoreactive for glycine, choline acetyltransferase, somatostatin, or tyrosine hydroxylase. To test whether CRABP‐positive somata in the ganglion cell layer were ganglion cells, we retrogradely labeled the ganglion cells of rat retinas with rhodamine dextran. No rhodamine labeled ganglion cells were positive with anti‐CRABP, indicating that the CRABP‐positive cells in the ganglion cell layer are probably amacrine cells. Other retinoid binding proteins thought to be involved in the visual cycle have been localized to Müller and retinal pigment epithelium cells. This study reveals that CRABP immunoreactivity is present in Müller and/or amacrine cells, depending on the species. CRABP immunoreactivity in amacrine cells of certain species was an unanticipated and provocative result. Its significance remains to be determined.