A role for the ciliary marginal zone in the melanopsin-dependent intrinsic pupillary light reflex

• First study of the iPLR in conscious wildtype and MKO mice.
• The iPLR in mice requires cholinergic neurotransmission.
• The iPLR requires signalling at the CMZ/ciliary body.
• There is a melanopsin-rich plexus at the nasal CMZ of mice.
• Retino-ciliary projections exist in mice and may contribute to the iPLR.

Maintenance of pupillary constriction in light-adapted rodents has traditionally been thought to involve a reflex between retina, brain and iris, with recent work identifying the melanopsin-expressing intrinsically photosensitive retinal ganglion cells (ipRGCs) as the major conduits for retinal input to the brain. There is also a less well-understood phenomenon whereby the iris of some mammals, including mice, will constrict to light when either the eye, or the iris itself is physically isolated from the brain. The intrinsic pupillary light reflex (iPLR) is the term given to pupil constriction in the absence of retinal input to the brain. Here, using an intraocular axotomy approach, we show that the iPLR in conscious mice spans a dynamic range over 3 log units of irradiance. This iPLR response is absent in melanopsin knockout (MKO) mice and can be significantly inhibited by atropine. Immunohistochemistry for cfos and melanopsin, in combination with light exposure revealed a population of small ipRGCs in the retinal ciliary marginal zone (CMZ), which remain responsive to light in axotomised mice. We report that damage to the CMZ in a novel in vitro preparation removes a significant component of the iPLR response, while a detailed immunohistochemical analysis of the CMZ in wildtype mice revealed a melanopsin-rich plexus, which was consistently most intense in nasal retina. There were clear examples of melanopsin-positive, direct retino-ciliary projections, which appear to emanate from Brn3b negative, M1 type ipRGCs. These cells are clustered along the melanopsin-rich plexus nasally and may channel ipRGC signals from retina into the iris via ciliary body. Comparison between wildtype and MKO mice reveals that the ciliary body is also weakly stained for melanopsin. Our results show that the full extent of iPLR in mice requires cholinergic neurotransmission and intact signalling at the CMZ/ciliary body. This response may be mediated to some extent by ipRGCs, which send direct projections from the retina into ciliary body. In addition to the melanopsin-mediated iris sphincter constriction suggested by others, we propose a new mechanism, which may involve constriction of the ciliary body and ipRGC-mediated relaxation of the iris dilator muscle.