Dark-adapted rod and cone system responses in a genotyped twin cohort: exploring heritability and associations with a myopia risk variant

Abstract

Purpose : Dark-adapted rod and cone responses can be measured by recording electroretinogram (ERG) responses to flashes delivered in the dark and in a dim blue background. We analysed these responses from healthy adult twins to investigate heritability and associations with a myopia risk variant.

Methods : ERGs were recorded with a conductive fibre electrode from volunteers from the TwinsUK cohort after mydriasis and 20 min dark adaptation. Following international standard scotopic stimuli, 4 xenon flashes were delivered (0.67, 3.9, 13, 67 photopic cd m^-2 s) in the dark, and then in a dim blue background (30 scotopic and 1.0 photopic cd m^-2) which saturated the rods but minimally desensitised the cones. Responses on the background were taken to represent the dark-adapted cone ERG; subtracting these from responses in the dark yielded estimation of the dark-adapted rod response. Coefficients for intra-pair correlation were calculated for monozygotic (MZ) and dizygotic (DZ) pairs. Amplitudes were compared between those homozygous for a myopia-associated risk allele (at locus rs52452) and those with no risk alleles. Where both twins of a pair were in the same group, parameters were averaged.

Results : ERGs from 208 participants (58 MZ and 46 DZ pairs) were analysed: mean (SD) age was 64.3 (11.5) years; 93% were female. Coefficients for intra-pair correlation ranged from 0.29 to 0.87 for MZ pairs and 0.11 to 0.55 for DZ pairs; correlations were higher for MZ pairs. Out of 148 participants genotyped at the locus of interest, 34 were homozygous for the risk allele and 35 had no risk alleles (with no significant difference in axial length). Of these, 30 and 27 respectively had reliable ERGs available for analysis. Amplitudes were significantly different (p<0.05) between groups for the cone b-wave elicited by the dimmest flash and cone a-wave responses to the other flashes. There were no significant amplitude differences for rod derived parameters.

Conclusions : Genetic factors are likely to be important in determining variance in both rod and cone driven responses as coefficients for intra-pair correlation were higher for MZ twins for all parameters. Also, the risk variant appears to be associated with differences in cone-driven signalling. These findings are consistent with alterations in cone pathways play a role in driving myopia.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.