Abstract
Purpose :
Mammals detect light for visual perception, but also for regulation of physiology, cognition, mood, and development. Many of these ‘non-image’ visual functions exhibit extensive integration of light over visual space. Intrinsically photosensitive retinal ganglion cells (ipRGCs), which capture photons with a receptor called melanopsin, are indispensable for many non-image visual functions. Melanopsin is distributed throughout ipRGCs, including in the axon. We hypothesized that phototransduction in axons allows ipRGCs to integrate light over an exceptionally large area.
Methods :
We performed electrophysiological recordings from ipRGCs in retinas isolated from transgenic reporter mice. We first determined if ipRGCs detect selective illumination of the distal axon, >1 mm from the soma and dendrites. Next, to measure the axonal contribution to overall photosensitivity, we compared responses elicited by illumination restricted to the soma and dendrites (local) to those elicited by illumination of the entire cell (global). Recording from ipRGCs in the central and peripheral retina permitted a natural comparison of axonal photosensitivity in cells with short and long axons, respectively. Antagonists of synaptic transmission were added to the extracellular medium to isolate melanopsin’s contribution to the light response.
Results :
Our observations were consistent with axonal photosensitivity. IpRGCs exhibited excitatory currents and spiking responses to illumination of the distal axon, but not to illumination of an equidistant location on the retina that did not include the axon. Peripheral ipRGCs (1.4-1.8 mm from the optic disk) were ~30% more sensitive to global than local illumination, while central ipRGCs (≤ 0.3 mm from the optic disk) were equally sensitive.
Conclusions :
Melanopsin phototransduction within the axons of ipRGCs generates electrical responses to photon absorption far from the cell’s soma and dendrites. The first steps of non-image vision produce greater spatial integration than previously understood.
This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.