Purchase this article with an account.
Ana Fakin, Valentina Cipriani, Stanley Lambertus, Nathalie Bax, Anthony G Robson, Kaoru Fujinami, John Chiang, Anthony T Moore, Michel Michaelides, Graham E Holder, Carel C B Hoyng, Andrew R Webster; Complex interaction and Hardy-Weinberg disequilibrium of ABCA4 disease-causing alleles provides insights into the pathogenesis of retinopathy. Invest. Ophthalmol. Vis. Sci. 2017;58(8):3423. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
A paradox exists between the prevalence of disease-associated ABCA4 alleles in affected patients on the one hand, and their relative ratios in the population on the other. We explore this phenomenon in two large patient populations and the ExAC database. We determine genotype-phenotype correlations.
Study included 397 probands from Moorfields Eye Hospital (MEH) and 166 probands from Radboud university medical center (Radboudumc) with a clinical presentation of ABCA4-retinopathy and two pathogenic ABCA4 alleles. The ratio of genotypes comprising null and missense alleles was determined and compared with the ratio expected according to the population data (Exac.broadinstitute.org). The analysis was repeated for 15 alleles (14 missense, 1 splicing) of known severity (Fakin et al, IOVS 2016). Phenotypes of 210 patients with two of the above and/or null alleles were studied with electrophysiology (ERG; departure from the null phenotype) and fundus autofluorescence (frequency of foveal sparing, FS).
Assuming Hardy-Weinberg equilibrium, the expected ratio of null/null, null/missense and missense/missense genotypes was 1:35:298, and the observed 1:6:7. The expected ratio of severe/severe, severe/mild and mild/mild genotypes was 1:20:128, and the observed 1:1.5:0.1. Specifically, there was a disequilibrium for two allele classes: (1)p.G863A, p.L2027F, c.5714+5G>A, and (2)p.G1961E: these occurred in homozygous, (1)/(1) or (2)/(2), but never in compound heterozygous state (1)/(2). Analysis of phenotypes suggested that the first retained higher ABCA4 function in foveal cones (high frequency of FS), whereas p.G1961E retained higher ABCA4 function in the extrafoveal retina (high frequency of normal ERG). Findings were replicated in the Radboudumc cohort.
The skewed genotype ratios results suggest incomplete penetrance for a large number of pathogenic ABCA4 alleles, depending on the nature of the second allele. Many people harboring two pathogenic ABCA4 alleles may thus be unaffected or have an unrecognized phenotype that does not lead to a referral to a specialist clinic. Further, allele classes that cause different phenotypes in homozygous state and do not occur in trans suggest that ABCA4 protein has at least two functions, which can be affected unequally and in some instances reciprocally compensated.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only