Abstract
Purpose: :
The phototransduction cascade in mammalian intrinsically photosensitive RGCs (ipRGCs) has yet to be fully elucidated. In addition to its requisite role in vertebrate rod/cone phototranduction, guanylate cyclase (GC)-generated cGMP has been implicated as an important secondary messenger in rhabdomeric Limulus ventral photoreceptors. To investigate whether GC-produced cGMP contributes to ipRGC photoreception, we examined circadian behavior and ipRGC light responses in double knock-out mice that lack both GC1 and GC2 (GC dKO).
Methods: :
The circadian rhythms of wheel running activities for GC dKO and C57B1/BJ control mice were monitored in 12:12 light/dark cycles, constant light and constant darkness. Co-localization of melanopsin and cFOS was determined in retinal tissue from light-treated and dark-maintained GC dKO and control mice. Isolated retinas were placed RGC side down onto an array of 60 extracellular electrodes, and recordings of light-stimulated (480 nm; 9.2 x 1014 photons/s/cm2) RGC spiking activity were obtained.
Results: :
GC dKO mice entrained to 12 h light/12 h dark cycles. There was a significantly greater (P = 0.0001) change in the period of circadian activity (Δ tau) in constant light, relative to that in constant dark, in GC dKO mice (1.96 ± 0.21 SD; n = 15) versus age-matched C57BL/6J control mice (1.52 ± 0.3 SD; n = 13). Light exposure (3 h) induced cFOS expression in 92.9% of counted melanopsin-positive RGCs in GC dKO mice, as compared to 71.9% in controls. Multielectrode array recordings confirmed retinas from GC dKO mice were light responsive. Upon exposure to 480 nm light, there was an increase in action potential firing that peaked, on average, 11 s after light onset (n = 158 units, 4 retinas; 1 s binning) which was substantially delayed relative to responses obtained from control retinas (peak within 1 s; n = 256 units, 4 retinas). Consistent with a melanopsin-based mechanism, the sluggish responses recorded from GC dKO retinas persisted in the presence of glutamate receptor antagonists and lacked an OFF component.
Conclusions: :
These results indicate that, in contrast to rod/cone photoreception, GC1 and GC2 are not critical for ipRGC photoreception. Somewhat paradoxically, the GC dKO mice exhibit circadian behavior that suggests the SCN is getting more photic input, a finding that also correlates with the increased light-evoked cFOS expression in melanopsin RGCs in the GC dKO mice.
Keywords: circadian rhythms • ganglion cells • photoreceptors