Antibodies to CaBP D-28K Labels DB3 Diffuse Cone Bipolar Cells in the Human Retina

Abstract

Abstract: : Purpose: To determine the distribution of the calcium-binding protein calbindin CaBP D-28K (CB) in the human retina. Methods: Three normal human retinas were obtained 6-8 h after the death of the donors and the removal of the corneas for transplant. The posterior eyecups were fixed in 4% paraformaldehyde in phosphate buffer for 24 h at 4° C; the retinas were dissected and processed for CB immunohistochemistry. Both flat-mounts and vertical sections were used. Some vertical sections were double-labeled with antiserum to choline acetyltransferase (ChAT). Results: CB immunoreactivity was localized to cones, perikarya in the inner nuclear layer (INL) and ganglion cell layer (GCL), and processes in both plexiform layers. Cones were labeled in their entirety and they formed a fairly regular mosaic in flat-mounts. In the INL and close to the OPL, some labeled perikarya had an axon descending into the IPL and these were identified as bipolar cells. The axon terminals of these CB-immunoreactive (IR) bipolar cells formed a narrow band of varicosities that stratified at the same level as the OFF-ChAT band in the IPL, suggesting that they belonged to DB3 cone bipolar cells. In flat-mounts, the morphology of these axon terminals in the IPL was similar to those of macaque DB3s. We could not positively identify the labeling of DB5s or other types of bipolar cells. The remaining CB-IR perikarya close to OPL lacked a visible descending axon suggesting that they were horizontal cells. Labeled perikarya close to the IPL were relatively small in size and were probably amacrine cells; some of them were double-labeled with ChAT antiserum. In the GCL, there were fewer CB-IR perikarya than in the INL. Most of them were double-labeled with ChAT antiserum. Since all ChAT-IR amacrine cells were double-labeled for CB, we concluded that all cholinergic amacrine cells, both in the INL and GCL contained CB. In the GCL, there were also a few large perikarya that contained CB but were ChAT-negative; these were probably ganglion cells. Conclusion: These results show that the localization of CB in the human retina is very similar to that reported to other primates. The labeling was robust even under conditions that were not optimal, and therefore this result indicates that CB would be a good marker to study several types of neurons in human retinas.

Keywords: 330 bipolar cells • 434 immunohistochemistry • 557 retina: proximal(bipolar, amacrine, and ganglion cells)