Abstract
Purpose:
Intrinsically photosensitive retinal ganglion cells (ipRGCs) can be activated by intrinsic melanopsin phototransduction as well as by extrinsic inputs from rods and cones. Studying melanopsin-based functions is challenging in humans, due to the difficulty in isolating melanopsin activation from rod and cone inputs. Here we introduced a five-primary photostimulator that can control the excitations of rods, S-, M-, L-cones and melanopsin at a constant light level to allow for assessment of melanopsin-, rod- or cone-mediated light adaptation.
Methods:
The photostimulator included a bundle of five optic fibers to transmit the lights from the five bright LEDs through a homogenizer and diffuser to achieve a homogeneous field. The lights from 5 LEDs (dominant wavelengths of 456 nm, 488 nm, 540 nm, 592 nm, and 632 nm) were controlled by a laboratory-created board using Pulse Width Modulation dimming. Using a silent substitution method, the photostimulator could generate photoreceptor-isolating stimuli that modulated the rod excitations alone (R), cone excitations alone (S-, M- or L), melanopsin excitation alone (I), or any combinations of the photoreceptor signals. We measured human pupil responses with the L, M, S, R, I, L+M, L+M+S, L+M+S+R, L+M+S+R+I modulations (16% contrast, 1 Hz, mean retinal illuminance of 2 Td-20000 Td in a one log unit step), using an EyeLink II Eyetracker (SR Research). For each condition, pupillary response amplitude and phase were derived from Fourier analysis.
Results:
For all modulation types, the pupil response amplitudes increased with increasing light levels, although the pupil responses were not measurable for the melanopsin modulation at 2 Td due to below melanopsin activation threshold, for the rod modulation at 2000 and 20000 Td due to saturation, and for the S- modulation at 2 and 20 Td due to low contrast sensitivity. The melanopsin-mediated pupil responses had similar phases to rods, L, M, L+M stimuli but was out of phase with the S-cone-mediated response. While the phases with the rod or cone-mediated responses were relatively stable over 2 Td -20000 Td, the phase of melanopsin-mediated responses increases with increasing light levels, suggesting speeding up of the melanopsin phototransduction.
Conclusions:
Melanopsin displays a unique light adaptation characteristic compared with rod- and cone-mediated light adaptation in pupil flicker responses.