I am doing a project on expression of GFAP in the zebrafish retina. GFAP is a marker for glial cells. I have found that the glial cells are more abundant the closer they are to the optic nerve at the middle of the eye. This may just be because the cells are more tightly compacted here than at the edges of the eye... But I am wondering if it may have something to do with the glial cells being involved with neuornal processes, and the optic nerve being the transmitter of signals from the eye to the brain?
Glia serve the classic support role in the retina, similar to that served by neuroglia elsewhere in the CNS. That role mostly consists of taking used neurotransmitters and recycling them such as in the glutamate-glutamine cycle. As Preece suggested, glial cells are a heterogenous group, and with a general marker like GFAP you could very well be observing two different types of glial cell, one with a uniform presentation and one specifically in the central retina.
That being said, the most familiar type of glial cell in the retina is the Muller glial cell. It is particularly interesting in zebrafish, where it has been shown to act as a neural progenitor cell upon tissue damage in adult animals As you can see from that article, it fits your requirement of being GFAP expressing, although I couldn't find any good flat mount pictures showing whether it was preferentially expressed in the central retina.
So that gives us two potential roles for glial cells in the zebrafish: as potential multipotent retinal stem cells and as housekeeping neurons present to clean things up. You could come up with reasons either of those might be important in the central retina, but for my money the arguments are all pretty hypothetical: "the central retina is more active and therefore has a higher requirement for removing neurotransmitter" or "the central retina needs to be protected from potential damage more than the peripheral".
One thing I think we have a good basis to rule out is that it's the optic nerve. From a retinal point of view, you refer to the point where the optic nerve passes through the rest of the retina as the optic nerve head or optic disc. Axons from ganglion cells all across the retina stretch out from their soma toward this single point. By the time they reach the optic nerve, the vast majority of axons are far from their axon hillock. That means that they're being propagated from one node of Ranvier to the next. Because that process requires voltage gated ion and not neurotransmitter gated channels, there is no neurotransmitter release associated with that process and consequently no additional reason for glial cells to be present to clean up.
Any particular glial cell you're noticing this with?
I know that Astrocytes modify Calcium concentrations in the synaptic cleft. Perhaps at places where there are more synapses, more Astrocytes must exist to regulate such thing? Here's an interesting paper on a possible role of Astrocytes on associative networks: