Aquaporin 4 and Kir4.1 Distribution in Retinal Glial Cells of Normal Macaque Eyes and Eyes with Experimental Glaucoma

Purpose: The goal of the study was to determine the distribution of aquaporin 4 (AQP4) and a potassium inward rectifying channel (Kir4.1) in the Müller glia and astrocytes of the normal macaque monkey retina and to see how the distribution changed in eyes with experimental glaucoma.

Methods: Five Rhesus macaque monkeys, ranging in age from 7 to 11 years, had experimental glaucoma induced in one eye by argon laser scarification of the drainage angle to restrict the outflow of aqueous humor from the anterior chamber; the fellow eye served as a control. Posterior segments from the experimental glaucoma and fellow control eyes were fixed in 4% paraformaldehyde with phosphate buffer. Retina samples were taken from several nerve fiber rich regions within 5 mm of the optic nerve head (ONH) in the vicinity of the standard optical coherence tomography (OCT) peripapillary scan. Samples were cut at a thickness of 40 µm on a vibratome and were subjected to antigen retrieval followed by exposure to a polyclonal antibody to AQP4, or to Kir4.1, a monoclonal antibody to glutamine synthetase (GS, a marker for Müller cells), and a polyclonal antibody to glial fibrillary acidic protein (GFAP, a marker for astrocytes and reactive Müller cells). Antibody binding in retina was visualized with the appropriate secondary antibody conjugate of fluorescein, Cy3 and Cy5. Images of immunolabeled retina were captured by confocal microscopy.

Results: In the retinas of control eyes, AQP4 immunolabeling was present in GS positive Müller cell processes most prominently in the outer plexiform layer (OPL), around blood vessels, their trunks surrounding the nerve fiber bundles, and vitreal endfeet. AQP4 labeling co-localized to a lesser extent with the GFAP positive astrocytes of the retinal nerve fiber layer (RNFL). However, in eyes with experimental glaucoma and RNFL loss, AQP4 was greatly reduced in Müller cells, whereas strong labeling was detected in GFAP positive astrocytes. AQP4 positive astrocyte processes aggregated in the region typically occupied by RNFL bundles and remained surrounded by Müller cell processes. In the retinas of control eyes, Kir4.1 immunolabeling was observed in Müller cells from OPL to RNFL, and was most prominent in the Müller cell processes in the OPL, the trunks surrounding nerve fiber bundles and processes around the blood vessels. Kir4.1 labeling in experimental eyes was similar to that in control eyes, except in the eyes with the severest loss of nerve bundles, where it aggregated in the region of the nerve fiber bundles.

Conclusions: In control eyes, AQP4 and Kir4 immunolabeling in the retina show a regional distribution similar to that described previously in rodents. The reduction of AQP4 in Müller cells of eyes with experimental glaucoma might impact water flux, particularly in inner retina. The increased presence of AQP4 in astrocytes of experimental eyes is a pathophysiological alteration that might reflect a compensation for the loss of AQP4 mediated water movement, or other changes in neighboring Müller cells and degenerating nerve fibers. The effects of experimental glaucoma on Kir4.1 channels were less obvious.