Association of IGF1 single-nucleotide polymorphisms with myopia in Chinese children

Introduction

Myopia is the most common refractive disorder worldwide, especially in Asian countries, such as China, Japan, and Singapore, and its prevalence is rising (Ang et al., 2019; Morgan, Ohno-Matsui & Saw, 2012). Holden et al. (2016) predicted that by the year 2050, almost half of the global population would be myopic, with 10% progressing to high myopia. High myopia could be closely associated with complications like glaucoma, cataracts, retinal detachment, macular hemorrhage, and choroidal neovascularization (Ikuno, 2017), which is one of the leading causes of blindness and results in financial burden (Iwase et al., 2006; Xu et al., 2006).

Myopia is a complex disease affected by both genetic and environmental factors, whose etiology and pathogenesis remain unclear (Morgan et al., 2018). Previous studies have identified 26 myopia-related loci (MYP1–MYP26) through various methods, such as family-based linkage analyses, genome-wide association studies, and whole exon sequencing studies (Cui et al., 2017; Guo et al., 2015; Wang et al., 2017; Xiao et al., 2016). Some candidate genes are associated with high myopia, while some are related to moderate myopia (Zhang, 2015). However, the genetic effect on myopia is an interactive process involving multi-genes and is heterogeneous among different populations.

In the MYP3 locus on the chromosome 12q23.2, the insulin-like growth factor 1 (IGF1) gene codes for IGF-1, which is similar to insulin in structure and takes part in many physiological process including aging, apoptosis, development, cellular growth, metabolism, protein translation, and differentiation (D’Mello et al., 1993; Werner & Le Roith, 2000). IGF-1 is a kind of single-chain peptide composed of 70 amino acids which was also known as somatomedin C (Savage, 2013). It plays an important role in human growth and development through GH-IGF-1 axis (growth hormone) (Furlanetto & Cara, 1986). IGF-1 is found to be expressed in most of the tissues and the locally produced IGF-1 may be more important than its circulating endocrine form produced by liver (Han et al., 1988; Le Roith et al., 2001; Yakar et al., 1999). During puberty and pre-puberty, which are critical times for refraction development, IGF-1 may also be the inner motivation for ocular growth. For instance, after an intravitreal injection of IGF-1, the axial length of chicks grew longer (Zhu & Wallman, 2009). In addition, the axial length of patients with Laron syndrome, a disease that causes congenital growth hormone resistance, became normal, to a degree, after IGF-1 supplementation (Bourla et al., 2006). These studies have suggested that IGF-1 could contribute to ocular growth and the progression of myopia.

Genetic studies have been used to genotype the single-nucleotide polymorphisms of IGF1 and have shown that it is significantly associated with high or extreme myopia in Caucasian and Chinese populations (Mak et al., 2012; Zhuang et al., 2012). However, these associations were not replicated in a Polish family cohort and in two Japanese studies (Miyake et al., 2013; Rydzanicz et al., 2011; Yoshida et al., 2013). Nevertheless, all previous studies focused on the relationship between single-nucleotide polymorphisms (SNPs) in IGF1 and high myopia, whilst the association of SNP in IGF1 with mild or moderate myopia, which usually occurs during puberty and pre-puberty, remains unknown. Here, we aimed to show that genetic variants of IGF1 contributed to the development of myopia in a young Chinese Han population.

Methods

Results

There were 281 myopes and 373 non-myopes with mean ages of 9.84 ± 1.55 years and 8.06 ± 1.43 years, respectively. There were no significant differences in gender between the groups. The mean SE of participants’ right eye was −2.55 ± 1.64 D in the myopia group and 0.84 ± 0.81 D in the non-myopia group (p < 0.001). The mean axial length of participants’ right eye was 24.58 ± 1.04 mm for the myopes and 23.02 ± 0.79 mm for the non-myopes (p < 0.001). No significant difference was observed in SE and axial length between right and left eyes (Table 3).

All six tagSNPs were successfully genotyped, and all of the duplicated samples showed the same genotypes with their corresponding ones. The genotype distributions were all in line with HWE in both the myopia and non-myopia groups (p > 0.05). Their alleles and frequencies have been shown in Table 4.

We compared differences in genotype between the two groups using four genetic models: the allele, dominant, recessive, and additive models (Table 5). The SNP rs2162679 showed a statistically significant association with myopia in the allele and additive models (p = 0.0474 and 0.04973, respectively). However, these associations did not exist after we conducted the 10,000 times permutation test. All the other SNPs were not significantly associated with myopia in these four models. After adjusting for age and gender, participants with the genotype AA in the SNP rs2162679 had a higher risk of myopia than those with the genotype GG (OR = 2.219, 95% CI [1.218–4.039], p = 0.009; Table 6), with post hoc power of 66.2% . Also, the genotype CC in rs12423791 and CT in rs5742612 were associated with a lower risk of myopia (OR = 0.407 and 0.648, respectively; 95% CI [0.185–0.896] and 0.443–0.948, respectively; p = 0.026) after adjusting for age and gender.

We performed a haplotype analysis between the myopia and non-myopia groups with a variable-sized slide-window strategy in Haploview software version 4.2 (Table 7, Fig. 1). The four-SNP window rs5742653 –rs2162679 had the lowest p value (Chi square = 5.768, p = 0.0163), and all SNP windows with p values less than 0.05 included the SNP rs2162679. However, no significant differences existed after the permutation test.

Discussion

This study investigated the association between the SNP rs2162679 in the intron of IGF1 and myopia in a young Chinese population during puberty and pre-puberty. Those with the genotype AA in rs2162679 had a higher risk of myopia than those with the genotype GG (OR = 2.219). The A >G variation in this position may protect against myopia. In anther word, individuals with G allele in rs2162679 of IGF1 have less chance of getting myopia. During puberty and pre-puberty, increasing levels of hormones, such as growth factor and IGF-1, accelerate ocular growth and the development of body stature. These genetic variants in IGF1 may be related to the onset or early development of myopia in young populations.

We found no significant association between rs6214, rs12423791, and myopia in this young Chinese population. However, within the MYP3 locus, these SNPs in IGF1 were previously associated with high myopia, a finding that has been replicated by several studies. Metlapally et al. (2010) were the first to test the SNPs in IGF1 and found that rs6214 was significantly related to any type of myopia, including high myopia, in Caucasian populations. However, this conclusion was not confirmed in a Polish family cohort or in two Chinese and two Japanese studies (Miyake et al., 2013; Rydzanicz et al., 2011; Yoshida et al., 2013). A 2017 meta-analysis including eight case-control studies published before June 2016 reported no significant association between high myopia and the SNP rs6214 in IGF1 (Zhang et al., 2017). Moreover, Zhuang et al. found that rs12423791, another SNP in IGF1, was associated with extreme myopia rather than high myopia in a Chinese population and that the haplotype containing this SNP showed a significant association with extreme myopia (Mak et al., 2012; Zhuang et al., 2012). However, Yoshida et al., Miyake et al., and a 2017 Chinese study failed to replicate these associations (Miyake et al., 2013; Wang et al., 2017; Yoshida et al., 2013). This may be due to ethnic differences or the severity of myopia.

Unlike the positive relationship found in our study, two previous Japanese studies found no significant associations between rs2162679 in IGF1 and high or extreme myopia (Miyake et al., 2013; Yoshida et al., 2013). Participants in our study were children in the puberty or pre-puberty stages, and most suffered from mild or moderate myopia. These findings suggested that rs2162679 was only associated with mild and moderate myopia rather than high or extreme myopia. To date, six MYPs (MYP6–10 and MYP14) have been discovered and mapped in populations with mild and moderate myopia (Cui et al., 2017). Baltimore et al. mapped a myopia-related region in chromosome 1p36 in an Ashkenazi Jewish population with a mean SE of −3.46 D (Wojciechowski et al., 2006). In addition, chromosome 22q12 was associated with moderate myopia in 44 large American Ashkenazi Jewish families (MSE = −4.67 D) (Stambolian et al., 2004). Even though myopia is a complex disease affected by environmental factors, and high myopia is thought to be hereditary, it is necessary consider genetic factors in mild and moderate myopia.

Though intravitreal injections of insulin or IGF-1 in chicks results in extreme axial extension, changes in the axial length in mammals are relatively limited (Feldkaemper, Neacsu & Schaeffel, 2009; Zhu & Wallman, 2009). This may be due to variation in their ocular structures and metabolism pathways (Mathis & Schaeffel, 2007). There is cartilage in the sclera of chick eyes, which could be stimulated by IGF-1 via cartilage anabolism (Martel-Pelletier et al., 1998; Rada, Thoft & Hassell, 1991). In addition, this structure could support axial elongation in chicks better than in mammals, whose eyes do not contain cartilage. This may explain why IGF-1 resulted in extreme increases in axial length in chicks but has only been related to mild to moderate myopia in humans.

Interestingly, rs2162679 of IGF1 had been reported to be associated with several kinds of cancer, in which the allele G could also decrease the risk of cancer onset (Xu et al., 2019). It reminded us that IGF-1 might play similar roles in myopia and cancer. More studies are needed to investigate the correlation between myopia and cancer.

However, this study had some limitations. First, the sample size was relatively small. Small sample sizes are more likely to result in false positives than larger sample sizes. They also have lower statistical power to detect positive results (Yoshida et al., 2013). Second, environmental factors, such as lack of outdoor activity or too much near work, are involved in the development of myopia, so we these may have confounded our analysis (Saw et al., 2000). Further studies with large sample sizes and in different populations are needed, especially those that stratify populations according to levels of myopia. Environmental factors and gene-environment interactions should also be taken into consideration. Last but not least, considering that rs2162679 locates in the intron region of IGF1, whether its variation could lead to protein function change remained unknown.

Conclusion

We found that the SNP rs2162679 in IGF1 was significantly associated with myopia in this young Chinese population, which provided clues for in-depth mechanism interpretation that IGF-1 may play a regulating role in the progression of myopia shift. Besides, it indicated that rs2162679 in IGF1 could be one potential genetic loci for myopic risk prediction.

Supplemental Information