DL-α-aminoadipate is a toxin to Müller cells

PEDF counteracts DL-α-aminoadipate toxicity and rescues gliotoxic damages in RPE-free chicken retinal explants

Gliotoxic responses complicate human eye diseases, the causes of which often remain obscure. Here, we activated Müller cells (MCs) by the gliotoxin DL-α-aminoadipate (AAA) and assayed possible protective effects by pigment epithelium-derived factor (PEDF) in RPE-free retinal explants of the E6 chick embryo. These models are suited to analyze gliotoxic reactions in vitro, since the avian retina contains only Müller cells (MCs) as glial components, and the RPE-free explants are devoid of a major PEDF source. ChAT- and AChE-immunohistochemistry (IHC) revealed that AAA treatment disrupted the differentiation of cholinergic amacrine cells in the inner plexiform layer. At the applied concentration of 1 mM AAA, apoptosis of MCs was slightly increased, as shown by TUNEL and caspase-3 activity assays. Concomitantly, cell-free gaps emerged in the middle of the retina, where MCs were swollen and amassed glutamine synthetase (shown by GS and Vimentin IHC). AAA treatment strongly activated MCs, as shown by GFAP IHC, and by an increase of stress-related catalase activity. Remarkably, nearly all effects of AAA on MCs were effectively counter-balanced by 50 ng/ml PEDF co-treatment, as also shown by RT-PCR. These findings suggest that supplementation with PEDF can protect the retina against gliotoxic attacks. Further studies should establish whether PEDF similarly protects a gliotoxic human retina.

Glial transcytosis of a photoreceptor-secreted signaling protein, retinoschisin

In vitro studies have clearly shown that signaling/guidance proteins can diffuse to their targets. However, it is unclear whether they can travel by diffusion in vivo, or if they are distributed in the tissue by an active mechanism. Retinoschisin, a signaling molecule related to neuropilins, is synthesized and secreted by photoreceptor cells in the outer retina; then it interacts with inner retinal cells contributing to synaptic organization and optic nerve fiber integrity. We developed an assay to examine how retinoschisin, which is secreted a distance away, reaches its inner retinal targets. We found that retinoschisin is preferentially taken up and carried into the inner retina from the retinal outer border (the photoreceptor side) by Müller cells (the main glial cells of the vertebrate retina). This transcytosis is disrupted by DL-alpha-aminoadipic acid, a Müller cell/glia-specific toxin. Our results suggest that glial uptake/transcytosis can provide an effective and precise alternative for distributing signaling molecules in the nervous system.

Glutathione depletion causes an uncoupling effect on retinal horizontal cells through oxidative stress

To investigate a physiological role of glutathione in the horizontal cells of carp retina, the gap junctional intercellular communication between horizontal cells was studied using the techniques of intracellular recording of light-induced responses and coupling of the fluorescence dye Lucifer Yellow. Intravitreal injection of 2.5 micromol L-buthionine sulfoximine, an inhibitor of glutathione synthesis, induced a dramatic reduction (20% of control) of retinal glutathione level two days after treatment. The low level of glutathione continued for a further four to five days, and thereafter gradually recovered to about 40% (20 days after injection) and 70% (50 days after injection) of the control level. The spatial properties of the photopic L-type horizontal cell response were examined by enlarging the diameter of the central spot and peripheral annulus over the recording point. In normal retinas, the response amplitude of horizontal cells was monotonically enhanced as the diameter of the spot increased (0.5-4.0 mm) and correspondingly the dye diffusion area was wide, as the injected Lucifer Yellow normally diffused to several neighboring cells. Treatment with L-buthionine sulfoximine significantly altered the spatial properties of horizontal cells by increasing the response amplitude to central spots and slightly decreasing that to peripheral annuli, which were observed by four days after injection. It also restricted intracellular Lucifer Yellow to one or two cells. Accompanying the recovery of the cellular level of glutathione, the spatial properties and dye coupling of horizontal cells were restored to normal. A time lag (two days) of initiation in retinal glutathione depletion and alteration of spatial or dye coupling properties of horizontal cells is discussed, together with reactive oxygen species accumulation.