Abstract
Purpose :
We recently reported in mouse retina that (i) connexin36 (Cx36) is required and sufficient for photoreceptor electrical coupling, (ii) every rod has electrical access to a cone, and (iii) rod/cone gap junctions are predominant in the photoreceptor network. Surprisingly, we have found no evidence for direct rod/rod coupling and cone/cone coupling is rare. Also, we have established the resting (dark-adapted) state of rod/cone coupling conductance (~ 300 pS) in B6 mice and its dynamic range [0-1200 pS] (Jin et al., 2020, 2022, Ishibashi et al., 2022). We developed a realistic computational model of the photoreceptor network based on our novel data to interrogate the impact of rod/cone coupling on the single-photon response (SPR).
Methods :
The model was built in MATLAB; a basic version of it has been published previously (Jin et al., 2015, 2020). The revised model includes 9 cones and 360 rods, with 40 rods connected to each rod. Importantly, we added uncorrelated current noise in each photoreceptor (Gaussian noise with SD = 0.5 pA), so that we could calculate the signal-to-noise ratio (S/N) in each photoreceptor of the network. S/N was computed as a function of light density or the fraction of rods eliciting a SPR (from 1 to 40, out of 40 connected to the same cone), and as a function of the rod/cone coupling conductance (from 0 to 1200 pS by 100 pS increments).
Results :
We estimated the S/N in uncoupled rods ~5. Increasing rod/cone coupling substantially decreased the voltage noise whereas it had moderate impact on the SPR amplitude, resulting in increased S/N for as long as at least half of the rods (20/40) elicited a simultaneous SPR. When less than half of the rods elicited a SPR, however, increasing coupling lowered the SPR more than the noise, resulting in smaller S/N. In the extreme case when only 1 rod (1/40) elicited a SPR, the S/N remained relatively unchanged when rod/cone coupling increased from 0 to up to 300 pS (resting state). However, the S/N decreased for coupling >300 pS, reaching a value of ~2.5 when rod/cone coupling was 1200 pS.
Conclusions :
The model reveals that the impact of electrical coupling on the rod SPR is much less than previously appreciated. Although rod/cone coupling decreases the SPR amplitude, the concomitant voltage noise reduction in the photoreceptor network conferred by coupling mitigates the effect of coupling on the S/N.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.