MYOPIA RESEARCH IN HUMANS
Gene finding:
First, we performed a meta-analysis of genome-wide association (GWA) studies within our international CREAM consortium (n=160,420). This led to the identification of 161 genetic loci for refractive error and myopia. All cells of the retina appeared to be sites of expression of these genes. Subsequently, we established an even larger study population in collaboration with the UKBiobank (n=542,934). This facilitated the finding of 336 more genetic loci. From these studies, we conclude that common refractive errors and myopia are caused by many genes with small effects, and genetic risk of the trait is caused by the sum of these genes.
We also sought to find genes running in families and in persons with extreme values of myopia. We found that about one fifth of these families and persons carry risk variants in hereditary retinal dystrophy genes, which is remarkable as these persons had not been diagnosed with diseases belonging to this spectrum.
Relation of genes to myopia severity:
We calculated polygenic risk scores (PRS, combined genetic effects) of refractive error in the Rotterdam Study. We found that persons with high myopia carried more genetic variants than persons with lower myopic errors, emmetropia (no glasses) or hyperopia (plus glasses). We also performed genetic studies (GWAS) in the UK BioBank study, and found that genes did not differ much between refractive error categories. These findings support the hypothesis that the whole range of refractive error is determined by the same set of variants, but the level of refractive error is determined by the genotype and number of genetic variants (genetic load).
Relation between myopia and glaucoma genes:
Myopia is a risk factor for glaucoma, it increases the risk of this trait 2-3x. We found no evidence to support a genetic overlap between myopia and glaucoma, suggesting that each trait is determined by a different set of common variants.
Environmental factors:
In our cohort study of Dutch children (Generation R), we investigated lifestyle habits. Our results confirmed the strong protective effect of outdoor exposure, and suggest that outdoor time should be at least 2 hours/day at elementary school. We developed the '20-20-2 rule' (after 20 minutes of near work, have a break for at least 20 seconds, and play outside for at least 2 hrs/day) for children, parents, and schools as a measure for myopia prevention.
Combined genetic and environment risks:
We studied the combined effects of genetic and environmental risk factors in children studies, and found that children with high myopia have higher risks of both environmental and genetic factors. We also combined risk factors into a polygenic risk score (PRS) and an environmental risk score (ERS) and found that the combination of PRS and ERS had the strongest influence on myopia development. We confirmed an association between parent and child myopia, but found that this association was determined by genetic as well as lifestyle factors. In other words, myopic parents create myopic environments for their children.
Prediction models:
We developed our prediction models in various study populations. We provided normative values for axial length from age 6 to adult age. These growth curves can be used to monitor eye growth and estimate the risk of (high) myopia in European children, and are currently implemented in commercial myopia meters (Visiaimaging). Using the risk profile parental myopia, books read per week, time spent reading, no participation in sports, non-European ethnicity, and less time spent outdoors, we could predict incident myopia with a predictive value of 78%. This did not require information from genes.
MYOPIA RESEARCH IN ANIMAL MODELS
Studies in mice:
We investigated the absence of a number of refractive error genes in mouse gene knockout models (Gjd2, Lrp2, Ano2, Nob, Pde11a) by functionally screening mutant mice strains. We tested refractive error, eye biometry and contrast sensitivity at different time points during the postnatal development. Susceptibility for myopia was tested using lens-induced myopia with negative powered lenses. Myopic phenotype was confirmed in a telecephalon specific Lrp2 knockout mice. Our studies showed that Gjd2 knockout mice have a lower bodyweight and relatively larger eyes. The lens showed opacities. These studies will guide new research lines to develop and test new therapies, such as scleral crosslinking.
Studies in zebrafish:
We generated knockout (KO) zebrafish models of 12 myopia-candidate genes to gain insights into the underlying myopia-causing disease mechanisms. We developed a myopia screening protocol that included quantification of refractive error, biometric measurements of the eye, visual behavior, and retinal function tests. Our studies in these models showed that absence of gjd2a leads to a hyperopic phenotype suggesting that up-regulation or increased coupling of Cx36 influences the risk of myopia. Furthermore, the study of other KO models (e.g. fbn1 and prss56 mutant fish) demonstrated that zebrafish is a suitable model to study refractive errors. To further investigate the effect of light exposure and to study gene x environment interactions we are now growing fish on different light conditions, and will assess the effect of light intensity and spectrum on eye growth. These experiments will provide additional clues on the mechanisms behind refractive errors.
