Genome-wide DNA hypermethylation and homocysteine increase a risk for myopia

Abstract

Aim: To test for the association between genome-wide methylation and myopia in human and mice.

Methods: Long interspersed nucleotide element 1 (LINE-1) methylation levels were used to surrogate genome-wide methylation level. We first tested for the association between high myopia (<-6 D) and LINE-1 methylation in leukocytes in 220 cases and 220 control subjects. Secondly, we validated the results of LINE-1 methylation in eyes from the form deprivation myopia (FDM) mice. Furthermore, we calculated the correlation of LINE-1 methylation levels between leukocyte DNA and ocular DNA in the mice. We also tested whether dopamine can alter LINE-1 methylation levels.

Results: The LINE-1 methylation level was significantly higher in the myopic human subjects than controls. The upper and middle tertiles of the methylation levels increased an approximately 2-fold (P≤0.002) risk for myopia than the lower tertile. Similarly, FDM mice had high LINE-1 methylation levels in the leukocyte, retina and sclera, and furthermore the methylation levels detected from these three tissues were significantly correlated. Immunohistochemical staining revealed higher levels of homocysteine and methionine in the rodent myopic eyes than normal eyes. Dopamine treatment to the cells reduced both LINE-1 methylation and DNA methyltransferase levels.

Conclusion: LINE-1 hypermethylation may be associated with high myopia in human and mice. Homocysteine and methionine are accumulated in myopic eyes, which may provide excess methyl group for genome-wide methylation.

Keywords: LINE-1; dopamine; homocysteine; methylation; myopia.

Figure 1. The mouse LINE-1 methylation levels in different tissues were correlated

The Spearman rank correlation of mouse LINE-1 methylation levels between blood and retina (A); blood and sclera (B) (n=11). The data from 2 eyes of each animal were averaged. The data were from normal control mice.

Figure 2. The LINE-1 methylation levels were increased in the blood and ocular tissues of FDM mice

LINE-1 methylation levels in blood (A), sclera (B) and retina (C) of FDM and normal control mice. FDM mice had significantly higher methylation level in the blood (P=0.027). The comparison between the covered myopic eyes and normal eyes yielded the P values of 0.018 for scleral DNA, and 0.032 for retinal DNA. For the comparison between uncovered fellow eyes and normal eyes, the P value was 0.021 for scleral DNA, and 0.023 for retinal DNA. Data are mean±SE. ^aP<0.05. FDM: Form derivate myopia (Data in this figure included 6 FDM mice and 11 normal mice. n=6 for covered myopic eyes; n=6 for fellow eyes. For the 11 normal control mice, the data from 2 eyes of each animal were averaged).

Figure 3. IHC staining in the eyes

Representative images of methionine staining (A), homocysteine staining (B) and negative control without staining (C). IHC staining showed increased levels of homocysteine and methionine in the retina of FDM eyes (n=2) than the data from the normal mice (n=2). No observable difference of IHC staining for either homocysteine or methionine in the sclera between FDM and normal eyes. Scale bar: 25 µm.

Figure 4. Dopamine reduced LINE-1 methylation level and DNMT1 expression

Treating RPE cells with dopamine for 24h caused significant reduction of LINE-1 methylation level (A) and DNMT1 gene expression (B). Data are mean±SE from 3 independent experiments. ^aP<0.05.

Figure 5. Schematic presentation for the relationship among myopia, methylation, homocysteine metabolism and dopamine

Dopamine can decrease DNMT1, which in turn reduces LINE-1 methylation level. Homocysteine and methionine are accumulated in myopic eyes, which may provide excess methyl group for DNA methylation.